A big hello to everyone around the world, wherever you happen to be tuning in to this live stream. Welcome, welcome to Cisco Live 2025 from beautiful San Diego. We are streaming straight to you in any language that you need, right here from the Cisco TV studio in the beautiful San Diego Convention Center. My name is Steve Moultrie. I am one of your five broadcast hosts. So good to have you back here on the broadcast with us. We have got an incredible three days ahead filled with keynotes and breakouts and customer stories and new technology announcements, deep dives, so much fun, so much industry leadership here on the broadcast. We will make sure that you do not miss out on a moment of it.
Across the world of solutions, right next door here in the sales pavilion, from the keynote stage to the incredible Cisco showcase that we're going to show you, it is all about our end-to-end, fully interconnected, fully secure one Cisco portfolio. We are combining the power of the network with security, with observability, with collaboration. That way we can help you, we can help your organization to thrive, to lead here in this AI era. The pace of change is overwhelming. Cisco has your back. We are your partner. Right now we are just under 30 minutes from our Tuesday opening keynote led by Chairman and CEO Chuck Robbins. I am joined here in the studio by one of our new Cisco TV broadcast hosts this year, the fabulous, the delightful, the amazing, the brilliant Zonafia Williams. Good morning, Z.
Good morning, Steve. I am super excited to be here this morning. I tell you, you know what excites me about being here is our presence, our presence as a company in this new AI era. I tell you, I was in the airport and I was not trying to ear hustle or anything like that. I am just going to say I had some intel and I overheard some folks talking. They were walking in the airport and these gentlemen were talking and they said, you know what? I want to find out what Cisco is doing, what they are going to be doing because they trust us. We are their trusted advisor. I tell you, Steve, I felt like, you know, when there was trouble in Gotham, you know, there was the Bat Signal and they were like, where is Batman?
I feel like people are saying, where is Cisco? What is Cisco going to do? I am super excited about what we are going to be doing, what is going to be coming across the stage. I am telling you, I am just elated to hear what is going to be next.
Your energy has been phenomenal since we arrived and started putting the production together. We're going to talk more about you because I want to know more about your history with Cisco as well, what you do with the organization. We're going to bring you in on all of that. Z, as we mentioned a couple of moments ago here, we are on a nonstop program. It runs all day today, tomorrow, Wednesday, Thursday. We're deep diving into investments and excitement and empowerment for all of our viewers. That's what you're talking about. We've got 24,000 people in town, like the people that you were hearing from, right?
Yeah.
Wherever people are on their journey, their data center journey, their AI journey, their journey to the cloud, whatever it happens to be, we are talking AI-ready data centers. We will talk future-proofed workplaces with these people. Powerful program will be there. They're Batman, right? Where is Batman?
Right there.
What I want to do before you and I keep talking is I want to go out into the keynote space to get our eyes and ears out there because things are starting to really ramp up. Oh, I have got some fabulous folks out there. Robb Boyd, the amazing Robb Boyd, who we adore, is there with Michelle Morera, another one of our amazing new hosts. Rob, let's send it out to the two of you.
Yeah, thank you so much, Steve. We're having a good time. Of course, you know, this is what we do, right? We're here to bring the energy. We've got so many people flooding in here. Hopefully there were some cutaway shots or something, but it's funny how you have a big room that suddenly gets a lot smaller as people crowd in here, but it's an excited crowd. I'm so excited to have Michelle with me. She's got friends all over the place. Michelle, what are your first impressions at this moment for Cisco Live?
This is my first Cisco Live and first keynote. This is what it's all about for me. I'm so excited to get to see what these Cisco execs talk about and bring to us today. I'm just overwhelmed by the dancing, the songs, the music, people coming in. It's just been incredible.
Yeah. So what we're going to do, Steve, we're going to be talking to some Cisco champions, a couple that were handpicked for the contributions and such that they've made. We're going to talk to them in a little bit. I know Lauren's out here. You'll introduce her in a moment. A lot of activity, but Steve, we'll take it back to you and enjoy the show.
you so much, Robb and Michelle. Welcome. It is so great to have you on our team. Michelle is just amazing. I want to right out of the gate mention to you social media, how important it is for all of you to stay connected to us here at the event. Hopefully next year you'll be right here in the room with us. For now, please connect with us on social media, whatever platform you prefer. Make sure that you include hashtags at Cisco Live and hashtag Cisco Live so that we can see it. We've got our social media team here on the show floor. They're going to respond to every one of your hits and every one of your photos and videos and comments, especially across the keynote, which we're going to introduce in just a moment. Keep reaching out to us.
We are going to meet the fifth member of our broadcast team right now, the amazing Lauren White, also out of the keynote space. Good morning, my friends.
Hello, hello. I am pumped. This is my ninth Cisco Live experience and fourth in-person event. I got to tell you, it feels amazing to be back. My joke is I worked at Cisco, loved every moment for almost nine years, found an exciting opportunity at this awesome company called Splunk. I guess Cisco needed me back so badly that they bought the entire organization to do it. Had no idea I was worth $28 billion, you know? In all seriousness, I am super excited to be here, to see the breadth and energy all around me. You know, I feel like I need to share this. I see the smiles, I hear the chatter. Steve, Z, do you feel that energy over there? Am I spreading it and am I pushing it? Do you get it over there?
You feel me sending it over there back to you guys? You feeling that?
I feel it. I feel all the energy coming out from you. Hang tight. Go find amazing people. We'll keep coming back out to the keynote space with you there, Lauren. Great, great, great job. Again, checking the clock, we are about 24 minutes away from the start of our Tuesday opening keynote here in San Diego. One of the hot topics that we're going to be hearing from the keynote stage all week long here at Cisco Live is the AI-ready infrastructure, right? Chuck, Jeetu, the entire team, they're going to do deep dives into that. I want to get a quick preview of that. I've got Jeremy Foster here in studio. How did you manage to make it up to the studio this early in the morning with all the action going on out there?
They said to be here. Always ready to talk about AI-ready data centers.
I love that. I love the command that we have here up in the TV space. All right. Jeremy, I want to kind of start at the foundational, right? Because there are a lot of people who are still a bit confused. How do we define AI-ready infrastructure? And what would you say maybe are some of the misconceptions people have, especially our customers, about getting to it and making it real?
Great point. I mean, there's two really big transitions that are happening within the data center right now, both related to AI. The first one is customers have all these traditional applications that are sitting inside their data center. They've been managing for the last decades or so. They need to be able to modernize that. That's really the first step is to take what you have, scrunch it down, modernize it so you can save power and cooling, and then use that rest of that space, power and cooling, to deploy your AI type of deployments.
There is a lot of new technology you're going to see, big announcements around AI in the data center, new hardware, new partnerships, and things that we're doing with NVIDIA that I think will really make a lot of people excited and help customers make it very simple and easy to deploy that AI step within the data center, which is really what our enterprise customers are asking us for.
Absolutely. I am so glad that you brought up the new announcements. We are going to hear a ton of them here in the keynote. I was down yesterday walking around the world of solutions. I saw the networking IT application there, 19 new capabilities just in that one demo alone. I think that is an important point because a lot of our customers, they have existing investments with us and they are thinking, okay, what do I do? Do I have to buy a ton more? Can I leverage what I have? Can I scale it? What do we say to them?
Yeah, absolutely. With these AI type of solutions, a lot of them are going to be driven by NVIDIA. They're the big name in AI, right? A lot of enterprise customers don't want to set up new silos for the next technology in a data center because operationally it's really painful, right? If I have to spin this plate and spin that plate and spin the plate, eventually one of those plates is going to fall down. A lot of those customers are Cisco customers. What our partnership with NVIDIA does is it allows them to use the same tools, things like Nexus Dashboard that they're already using today, and extend that over to the AI portions of their new deployments. It's a great operational way for them to take control of their environments.
Is that kind of what we want to tell people in terms of the story about how we go from pilot to full activation on this is utilize the partnerships, utilize the people that we already work with who have already proven their value?
Yeah, that's a big piece of it. And utilize the solutions, right? So what we're doing with things like NVIDIA's AI data platform is taking our hardware from a compute perspective, network perspective, putting it all together into a validated design and even working with third parties like Vast Data, for example, or Pure Storage, so that they have an end-to-end reference architecture that they can leverage to make those deployments easy on day zero, but then more importantly, really easy to operate over the next five years of the usable life of that equipment.
I love that. We are there to empower people. I think a lot of times they think they have to do it alone. How is Cisco helping our customers to modernize without having to totally start over again, right? Let's say that people have been invested in one particular area or one particular direction and now they're thinking, well, I got to go this entirely different path. They don't have to, right?
That's right. From an AI perspective, obviously a lot of people are just getting started, but your question around modernization is a good one because they're going to have a mix of our compute, somebody else's compute, hopefully a lot of Cisco networking. It is about those operational tools and things like Intersight we can do with UCS where we can help bring these things together so they can deploy systems very quickly and scale very effectively without having to rework because we're adding a ton of automation within these platforms, which is some things you'll see in demos on the floor out there as well.
It's just how simply and easily we can help a customer get an environment up, leveraging templates, policies, and profiles so that they don't make mistakes, they don't build snowflakes that are going to be hard to worry about on day 365 that end up kind of biting you when you have infrastructure that's not configured the way you expect it to be.
Absolutely. All right. Last thing that I want to ask you, when we're here on the broadcast, it's all about get here, get to the event, be here in person, be here with us, get a chance to talk to Jeremy Foster. What does your week look like? Who are you talking to? What are the stories that you're talking about? What are you hearing from customers, from partners when you're here in the show space that would tell somebody else, now you got to get here and have these one-on-ones?
Yeah. I mean, a lot of our enterprise customers are saying, hey, help us make it easy, reduce our time to value, make it simple to do these deployments in the AI space. At the same time, there's a ton of conversation about, hey, my data center is not ready for AI. How do we go through that modernization story with them? The partners, right? I had a great breakfast yesterday with a ton of partners. It's awesome to get their feedback and see their excitement around what we're doing in the data center because they can really see that the investments that we've been making over the last couple of years are paying off and they love what they're seeing out in the showroom floor as well.
This is exactly why we love Cisco Live. Jeremy, I am so glad that we got you here and that we got you in early before you went haywire and before everybody grabbed up every moment for attention. Really glad to have you here with us and thanks for the amazing work that you're doing.
Thanks for having me. I appreciate it.
Truly appreciate it. All right. We are going to go back out to the show floor where our own fabulous Lauren White is hanging out with Mark Patterson. Lauren, let's head out to you.
I am here with Mark Patterson, Executive Vice President and Chief Strategy Officer and incoming Chief Financial Officer, Mark.
That's a mouthful.
It's amazing though. It's incredible. This week we have tens of thousands of attendees at Cisco Live. From your perspective, what makes this event such a powerful event for our customers and partners all around the world?
Yeah, Lauren, thanks for having me. I've been at Cisco now 25 years, been to countless Cisco Lives, and this is always an amazing event. The energy in here is always at a high. I think this particular event though, this year, there's so much innovation coming. We are the company that's building the critical infrastructure for the AI era and the agentic AI era. This is going to be an amazing event.
You know what? Speaking of AI, I couldn't agree more. It's been a transformative year for us here. How would you describe Cisco's progress we've made over the last 12 months? How can we really harness that momentum for our customers?
It's a great question. You know, there's been so much progress this past year. If you really just think back a year ago and turn back the clock, a lot of people didn't know what our role was. What is Cisco's role in AI? Now you fast forward, you've seen the momentum that we've had in the hyperscale space, exceeding our billion-dollar target a quarter early. You saw a bunch of announcements that we made around NeoCloud and Sovereign AI buildouts. We're now talking a lot about, we're talking about a lot of stuff here the next couple of days, but we're going to be talking a lot about the enterprise and agentic AI and the kind of technology, the way we will actually build the networks and secure AI going forward. It's going to be pretty amazing.
Amazing, amazing. I'm pumped. I can speak for everybody else too, I'm sure. When I made the introduction, you said it was a mouthful, but it was like a huge, huge just news, right? It was recently announced that you'll be stepping into this role come FY 2026, the role of CFO. How would you, like, what's been most energizing about this transition so far? What can you say about how that will impact Cisco's growth and transformation from that lens?
Yeah, I'm really excited, first off. I mean, this moment in time to become the CFO for Cisco is a huge honor. A lot of people don't realize this, but I came into Cisco from an acquisition where I'd led finance and operations and originally a CPA. This is a bit of a homecoming for me. I spent time in ops, finance, strategy, 11 years in the Salesforce, excited about coming back into finance. I think right now this is a moment where, honestly, in 25 years, I'm not sure that I've seen us have this much opportunity ahead for us. Now we got to execute, but it's going to be exciting times for sure.
I cannot wait. I feel like there's never been a better time to be a Cisconian right now. If you're not at Cisco Live, where are you? Thank you. Thank you so much, Mark, for your time.
Thank you, Lauren. Appreciate it.
Sending it back over there to the studio.
Thank you so much, Lauren. I appreciate it. You just said it exactly right. If there's never been a better time to be here, part of Cisco, part of Cisco Live. Congratulations to Mark as well on this great change for him and the leadership that he's going to bring. Really, really do appreciate that. All right, Z, I'm checking the shot clock. We're about 14 and a half minutes out from going to the opening keynote. I'm super excited about that, right? We're going to see Chuck take the stage in just a couple of minutes here. I want to kind of shift gears a little bit now because you have just joined the team. I want to talk about your area of specialty. Program Manager for Inclusion Collaboration here at Cisco.
I want you to talk a little bit about that core focus here at Cisco from your day-to-day, but then how does it play out to our team members, to the people that we work with, and all the way to an event like this?
Absolutely, Steve. You know what drives my or fuels my cup, you know, being in this space? I've been with Cisco since 2015, and I, like Michelle, we're twin souls. This is my first Cisco Live. I've watched it, you know, from a virtual perspective, but seeing it from this live lens, I mean, it's spectacular. I tell you, being in the social impact and inclusion space, it really, it speaks to our purpose here at Cisco. Our purpose is to drive an inclusive future for all. I think that translates into our people. It translates into our technology. You see the breadth of it when you come into this space here at Cisco. People, we are the trusted advisor. People trust us. Because they trust us, it brings a humanness to what we do. That's what really fuels me.
I love being in that area because it's about our people. You know, you can't have a great product if you don't have great people. I think it speaks to that. I think you feel that, you see that all around. Whenever you're around a Cisconian, you know you're in good company, Steven. I love just being here. I'm super excited about.
It just pours.
Being excited about it.
Yeah, why this is so important, why this is so valuable to you personally, it's an internal thing, right? It's a personal passion. You just hit on something very important that I think we're going to come back to again and again. This event is about technology, right? We're finding out the new technologies, we're discovering what's going on, what is Cisco developing, why is Cisco developing it based on demand that we've heard from the marketplace and from the customer, but fundamentally it's about getting people together in a room to have the conversations that really matter. Z, thank you. I'm going to keep coming back to you again and again with us. Right now though, we're going to go back out to the show floor where Robb and Michelle are hanging out with a couple of our fabulous Cisco Champions.
Oh, absolutely, Steve. I think you remember this, but I am also a Cisco Champion. Something I enjoyed being able to do about five years ago when I was freed up from direct Cisco overlordness. Either way, I have so much fun with this group because these are the kind of people that I learn from on a regular basis. In fact, these are two of the most involved people, I should say, because not only are these Cisco Champions, they're NetVets. Jonathan and as well, I mean, sorry, Jonathan and then Jeremy. Tell us real quick, Jeremy, where are you from and what do you do?
I'm from Wyoming Machinery Company, middle of nowhere, Wyoming. I'm the network admin out there for 21 years this year.
Wow. How many Cisco Lives have you been to?
This is my 10th Cisco Live.
10th Cisco Live. Okay. We'll see if another number holds up here. Jonathan, how many Cisco Lives have you been to?
This is number nine for me. I'm one year behind.
All right. But you two are working together is what I see here because he's got you beat on the additional ribbons and such here. Also, I want to make sure, of course, you've met Michelle. She's our newest co-host and she's doing her thing. We wanted to get some tips from you guys because of your experience. Before we do that, I'm just remembering, Jonathan, tell us who you are and where you're from. What do you do?
Yeah, my name is Jonathan. I'm from Iceland, a network engineer and a consultant focusing on security-related network products.
Perfect. Any hot tips from you?
Like for Cisco Live, fully utilize being in person. It's a walk-in labs, meet the engineer, and everything that you can actually do on site is my priority for the, yeah, being on site.
Excellent.
I like that as well. Jeremy, what would you say is a good tip for anyone that's not familiar with how to get the most out of the experience here?
Become a Cisco Insider. They do have the user groups where you get in on the roadmap sessions, and those are definitely not recorded here. You have to get in on the NDA. Those are definitely something to hit.
Yeah. The Cisco Champion program often involves you guys in some information that's not always shared publicly because you get a chance to kind of reflect back to product managers and developers. Have either of you been involved in those processes? Start with you.
Yeah. I've been part of some of the beta programs and the feedback sessions because I want to get the feedback across and make the product and the solutions better. Coming from Iceland, it's a small market, so making sure that my market's voice is being heard.
That's awesome. I know you're telling me you do a lot of things both with the insider team and almost like you're helping us build the infrastructure to make for better insider events and interactions. It feels like you're very much involved in the community. What would you cite as some of the things you've enjoyed getting involved in?
Probably the feedback. Being able to steer Cisco, saying, "Hey, this is what the customer wants." Being able to give that feedback and seeing it come to life in all the products, it's really helpful.
That's awesome. I want to thank you guys for what you do with the Champion program. Big shout out to Danielle. She runs this whole thing and she's always, this is a very, very busy week for her, herding you guys because these are just two of a big, there she is, big group of nerds. My nerds. I can't tell you how many times I go to this group asking for, "What should I be thinking about this? What's the non-Cisco opinion of something?" That type of thing. I love this. Thank you for what you're doing. Michelle, thank you. Steve, we'll throw it back to you in the studio.
Rob, so many important points. Thank you for being a champion as well, Rob. You know what? We just saw there, and we're going to see this again and again throughout the course of the week, Z, the champions are our feedback. They are eyes and ears out there in the marketplace. We just heard from the champion who had, by the way, the longest list of records I've ever seen hanging off of his badge. The whole idea that we want to make sure that our market is being heard, that what people are saying to us here at Cisco is, "This is what we need you to do. This is what we need you to create." That leads to not only our product developments, it leads to our partnerships. It leads to us looking around and saying, "Okay, where are our weak spots?
Where are the gaps that we can fill in?" We go out and we make that outreach to them. I think that's an incredibly exciting part of what it is that we do. Yeah. What I'd love to do, I want to do a very, very quick preview now of what's coming up because if I check the clock, we're about eight minutes away at this point from heading into this exciting keynote space and onto the stage. This morning's theme, Vision for the Future. Cisco Chair and CEO Chuck Robbins is leading an all-star deep dive into Cisco's groundbreaking innovations to connect, to protect all aspects of your organization here in this AI era. The team is really going to hone in today on AI-ready data centers, future-proofed workplaces on Cisco's unparalleled digital resilience.
Chuck is going to welcome Jeetu Patel, President and Chief Product Officer. Carrie Palin is on stage this morning, Cisco Senior VP and Chief Marketing Officer. DJ Sampath is going to join them, Senior VP, AI Software and Platform. We are going to meet some special guests as well. Alan Rosa is with us, Chief Information Security Officer and Senior VP of Infrastructure and Operations at CVS Health. We've got Kevin Weil on the stage this morning, Chief Product Officer at OpenAI. They are going to help show exactly how Cisco is harnessing AI-powered capabilities to modernize our infrastructures. Why is that important? Faster, smarter, more secure workplaces, more opportunities. Cisco is bringing the power of the network together with security, with observability, with collaboration. Cisco changes the game in the data center, the workplace, how every bit of it all comes together and gets protected.
Z, I want to ask you, you had a chance, six minutes, six and a half minutes, by the way, until we hit that keynote. I should have mentioned that. You had a good opportunity. I was locked here in the studio, like I always am, but you had a great opportunity to walk around yesterday, see what was happening here in the sales pavilion, down in the world of solutions. What personally kind of touched you where you said, "That's why we do this"?
Wow, Steve, I tell you, wow, there's so much to see down there. I tell you, they have some amazing relaxation chairs, but I tell you, the one thing that stood out to me, again, connecting back to the people, we didn't just bring thousands of people into the city just to consume space.
Or sit in relaxation chairs.
Or sit in relaxation chairs. I mean, what do you do? What do you do? We have an amazing opportunity to connect with this community right here in San Diego. We have a give-back opportunity where we are asking all of the attendees to contribute to those that were impacted by the LA wildfires. We have partnered with an amazing organization where you can make some packets and give back. It's all about connecting people to the heart of what's important to them. I tell you, that's what's super exciting about being here at Cisco Live. That was one of the key things that really stood out for me, Steve.
I love that that stood out. You know what else I love? Lauren is out there dancing for us in the keynote space. When Lauren starts dancing, you know you got to send it over to her. Lauren, what are you dancing about? Talk to me.
You know what? I am here with the incredible Marin, who is part of one of our most elite communities at Cisco, the CCIE NetVets. And she's over here getting me pumped. She started dancing, so I started dancing. Marin, tell us more about, you know, how long have you been a CCIE? How many Cisco Lives have you been to? Let us know.
All right. This is my 10th year as a CCIE, which is amazing. I used to teach, and there are not enough women in IT, let alone being a CCIE. I am always about that. My first Cisco Live was 2009.
Wow.
I remember how overwhelming it was because there are so many people and there's so much to do and so much to learn. I come most years. It's just fun. It's so fun.
I love your energy. I love your excitement. What are you most looking forward to hearing about this year at Cisco Live?
I know everybody's all about AI this year, but my first session this year was on Monday morning was about quantum computing. And that is like really tickling my brain. I'm very excited by this.
I am too. I need to actually meet you in one of those sessions. Maybe we can go together.
All right. I like it.
We're ready. We're pumped. You see everybody around us. Thank you so much, Marin.
It was a pleasure. Thank you.
Nice meeting you. We're getting ready for the session. We're going to send it back over there to you guys, Steve and Z.
I appreciate that so much. Thank you, Lauren. Thank you again to Robb and Michelle for going out and finding amazing people out there on that show floor. Three minutes and 45 seconds away, we are getting excited. We can feel it up here in the studio. We can feel it across the entire conference. They just said it a moment ago down there. The two of them were talking about, first of all, not enough women in technology. We can go ahead. We can talk. We could do an entire broadcast and an entire podcast on that. We do. AI, AI, AI, the barrage. We see what happens with people. They sort of get a wash. Yes, it's an imperative. Yes, we have to do AI.
Yes, we need to incorporate it into what it is that we're doing across our organizations in our security, our IT, our networking into the data center. People get a bit overwhelmed, right, in terms of the demand. How do they apply it not only securely, but ethically, morally? How do we really get maximum benefit out of it? Again, it's one of those details that we're going to dive into throughout the course of the week. Before we do that, I want to go back one more time out to the keynote floor where Robb Boyd has tracked somebody fun down. Rob, who do you have with you?
We've got a guest here. Michelle, do you mind introducing our guest?
I'm going to introduce the guest. Please tell us who you are and where you're from.
Hi, my name is Rolin Carris. I'm from the city of Idaho Falls. I am a system administrator there. I've been there for almost two years.
Excellent. How many Cisco Lives have you been to?
This is my first one.
Me as well.
This is a good impression time, right?
I'm curious. When you came in here, what expectations did you have? Forget what you've already learned. Did you have certain expectations about what to get at an event like this?
I knew that it was a really big event, and I wasn't quite sure exactly what I would learn or who I would meet. One of my expectations was I was looking to advance my understanding of Cisco, the products that we use, and how to integrate them together so that we get better coverage and better management under a single pane of glass because we have some disparate systems that I'm trying to look to bring together.
Yeah, that's got to be a common issue is you're trying to get how do we get that look together. I apologize. We're going to have to go back to the studio, but I want to thank you for coming here. I hope your Cisco Live goes well for you and that you come back for more. All right. With that, Steve, we'll go back to you.
Rock and roll. Thank you so much again, Rob, Michelle, Lauren, being our eyes and ears down there. We can see down here on our own monitor, Z, how things are loaded up. It's an incredible look. I mean, it's nice and quiet up here, but for boy, what's happening down there on the show floor is just so incredible. I hope you are as excited as I am. You ready to get going with this?
I'm ready to get going, Steve.
All right. Let's go ahead and launch onto this. I'm going to make the final pitch to everybody out here who's watching around the world so that we know exactly what we're headed for. Such an exciting morning ahead. The time has come. We are about a minute away from heading into the opening keynote Vision for the Future. Chair and CEO Chuck Robbins, Jeetu Patel, Carrie Palin, DJ Sampath, special guests from OpenAI and from CVS Health. We are going to be hearing multiple new product announcements, as we talked about with Jeremy, everything protecting and connecting every element of our organization in this dynamic, constantly shifting AI era. Key stories, opportunities, right, coming to you.
Everything across the workplace, digital resilience, all showing why Cisco is the partner for you to help you modernize your infrastructure, to help you deliver faster and smarter and more secure outcomes, and all the different ways that one Cisco is bringing the power of the network together. What does that mean? It means security. It means collaboration. It means networking, putting all those pieces together into a single value-driven, trustworthy platform that you do not have to do by yourself. Count on us here at Cisco to partner with you, to help you every step of the way. No matter where you are in your journey, we will get you where you want to go. I want to remind you, keep posting on social media, on Insta, LinkedIn, Facebook, whatever you like, whatever gets you excited, inspires you.
You hear something from the keynote stage and you say, "There." That is why I want to be a part of Cisco Live, why I want to be in the room live, or why I want to connect in. Also, feel free to go ahead and type something out that says, "Here's what I would really love to hear from the keynote stage, from one of the executives, maybe during an interview right here in the studio." Our social media team is out there reading, paying attention to, watching every single one of your posts. Just make sure that you include #CiscoLive or @CiscoLive. Better yet, go ahead and put both in there. It's just safer that way. That way they can read them. They will respond to every one of those messages. Wherever you happen to be watching from, go ahead and send us a photo of that.
We want to see you in your workspace or in that coffee shop or sitting on that sofa, whatever it happens to be. We want to see where you're watching and experience it all along with you, right? Add those photos, add those snaps. We are 10 seconds out. So excited about the opening keynote. Get out there, enjoy it. Keep posting on social media. We are going to meet you right back here in the studio when it's over. Away we go.
What's next? Starts now. Together. Connections spark innovation. Ideas ignite breakthroughs. Technology fuels possibility. Let's discover more. Expand our horizons. Transform what's next. The brightest minds. The biggest leaps. The boldest change makers. Let's design the future together. This is power. This is community. This is Cisco Live.
Please welcome Senior Vice President and Chief Marketing Officer, Carrie Palin.
Good morning, y'all. Good morning. Yes. We have 22,000 of you with us this week in San Diego. We do not take for granted that each of you chose to spend time with us and commune with us and learn this week. We have curated an epic lineup for you. I'm so excited to talk you through it. Today, we're going to start by bringing up our Chair and CEO, Chuck Robbins, to talk about how Cisco is well positioned to be your partner on your AI journey. After that, we will welcome to the stage our new President and Chief Product Officer, Jeetu Patel, not brand new, but a year into the role. You will be so amazed by the incredible innovation Jeetu and his team are driving on behalf of all of you.
He will welcome to the stage a very special guest, the CISO of CVS Health, Alan Rosa. Yeah. They are going to talk about the Cisco AI advantage and the power of the portfolio. He will end, and you do not want to miss this. I want to stress, do not leave because the end is the big drop the mic because Jeetu's going to have our brand new board member and Chief Product Officer of OpenAI, Kevin Weil, to talk about agentic AI. Folks, if you've never heard Kevin, you don't want to miss this. Yeah. It'll be bonkers. When you're done with keynote, make sure to stop by the World of Solutions, where you can get hands-on with our amazing technology and our amazing partners and have a whole lot of fun communing with your peers.
Lastly, make sure you stop by the Cisco store. We have a limited edition hoodie this year, only 150 of them, and also some amazing merch with Cisco and McLaren Racing collabs. Go get yours and have some fun. With that, I'm going to welcome to the stage Chuck Robbins, Chair and CEO of Cisco.
Limited edition hoodies. We've really arrived now. You know, the watch company sells scarcity. The car company sells scarcity. We're bringing scarcity to Cisco Live, only 150. Don't leave now, though. They'll be there afterwards. Welcome to everybody. We're so glad you're here. For all of you in the room, thanks for joining us. For everybody in the other rooms here in San Diego, thank you for being here. For everybody watching remotely, we're glad you're with us as well. Now, my team had in my notes today that I'm supposed to tell you that I'm excited. I'm excited. They also wanted me to say that this is perhaps the most important Cisco Live I've ever been to. It made me think. Every quarter, I tell my team, this is the most important quarter we've ever had.
I thought, do we say that every year, that this is the most important Cisco Live ever? We probably do. This year, we mean it. I'm serious about that. If you saw all the press releases that hit this morning, which most of you probably have not yet, the payload of innovation that you're going to hear about today is probably the most that we've delivered here at Cisco Live in my career. Hopefully you're excited. I think it is perhaps the most consequential time for us to come together, given what's going on in the industry. It made me reflect back in the late 1990s. Yes, I'm old. With the internet build-out, it was incredible. This is moving faster. It's more complicated. The upside is even greater than what we saw then.
Everything about what we're doing today starts with a very cool technology. It starts with this technology that we've all been around for a very long time, the network. We started as networking companies. This event used to be called Net Workers. Thank you. There are 12 people out there that still remember. We believe, and most of you have told us, that the network is more important as we make this transition to agentic AI than it ever has been. All the innovation that you see this week would not occur without the network. It is dependent upon the network. Our opportunity is to work together in this AI era. It will be, for all of us, the most important work we've done to date. We are going to talk a lot about one particular convergence that we believe is going to be critical for agentic AI.
That is the fusion of security into the network. Fusing security into the network. You're going to hear a lot about that. The race is on for agentic AI. The pace, as we all know, is unbelievable. Perhaps we can all right. We have a really good team because I was prepared for that. As I was saying, the race is on, and we all know it. The pace is unbelievable. What we're learning is unbelievable. When you hear Kevin later, he's going to talk about things that they planned on doing maybe two months ago. They've already changed or they've already evolved and already moved to the next thing. Last year, we were talking about chatbots and inquiries and efficiencies. Now we're talking about how do we manage the workforce, humans, and agents. 98% of you feel an extreme urgency to deliver on AI.
Eighty-five percent of you think within 18 months, you have to do it within 18 months. Very few of you feel truly prepared. This is a global AI race. I think we're all aware of that. I was asked yesterday at an analyst event, what's the biggest concern I have? It's the geopolitical dynamics that we're all facing today. We all know that every company, every organization, every government around the world, there's also a defensive mechanism associated with this. There's a FOMO element. Your CEOs, I talk to them. They want to move fast. They're worried about safety and security, but they're also just as worried about having one of their competitors move faster and create a competitive advantage that they don't have.
I suspect that every one of you either presents to the board at every board meeting or presents material that's presented at the board, if you have a board, whatever your governing structure is, about what are you doing in AI. I would contend that your role is more important than ever. How you run, how you secure networks, the demands that you're dealing with, computing power, bandwidth. We're going to talk a lot about the bandwidth requirements of agentic applications and agent-to-agent communication and the real-time low-latency need to transmit those requests and those responses. Lower power consumption, reimagine security. This creates a great opportunity. It creates a great challenge. It creates very strong pressure on all of us, which is kind of healthy, I hope. The bad news is you're all at the heart of making this happen.
The good news is you're at the heart of making this happen. I remember I was in a meeting a few years back, and somebody made a comment and said, "What if no customer ever buys another Ethernet port?" I looked at him like he was crazy. He was crazy. We're collectively at the heart of driving this change. Secure networks are going to be at the heart of it. 97% of you said modern networks are critical to make this happen. There are three big things we know. The network's foundational. Hopefully, we've established that. The second interesting thing that we see is that private data centers are back. Boy, I could talk forever about all the pressure we felt on the transition to cloud.
I think that was the same time when someone said, "Why will anybody ever buy an Ethernet port again?" Now most of you are looking at a very balanced hybrid approach and thinking through clearly you're going to use the cloud. There's great efficiency, and there's great, great value, as you've seen over the years with our wonderful cloud partners and what they do for all of us. You're also going to have smaller instances, smaller models. You're going to have small models running at the edge. You're going to need to work hard on building out your own private infrastructure to go with your public infrastructure. We need a new approach to safety and security. Most everybody in here has probably experienced some sort of security incident in the last 12 months. Here's an interesting stat.
I think it was last year, less than half of you believed that your employees in the organization you're in actually understood the risks associated with AI. That was before we started moving towards agentic AI. We need to fuse security into the underlying network in order to be able to apply network services, security services at wire speed while you're running these applications. We believe that we have a meaningful role to play. We have networking, and we have security. None of our networking friends have security, and none of our security friends have networking. It's an interesting place to be. I think we have the right portfolio. This week's an example of the pace of the innovation that our teams are delivering.
I really believe that if we establish the fact that we need to put security into the network in order to deal with agentic AI applications, then we really can't build secure AI without what we can do together as we leave here. I don't believe there's any company better suited right now to help you. Now, you might say I'm biased, but I like to think I'm objective. There are lots of companies that are going to help you. OpenAI is here, NVIDIA, AMD, all. There's lots of companies that are going to have a role to play. We think when it comes to secure networking, the fact that we have networking, security, and silicon, don't underestimate the combination of those: secure AI, AI-powered collaboration, the platform advantage, and don't underestimate the power of programmable silicon as we build platforms as we go forward.
If you run into Martin Lund around here, he runs our silicon business, and he's walking around with our latest chip in his back pocket. You can have him show it to you. We have a long history of working together with all of you. Many of you have been coming to this event for a very long time. We're proud of what we've done together. We're proud of the trust that we have in each other, both our customers and our partners who are actually attending this event, and the global partnerships and the ecosystem. Hopefully, you've seen how active we've been with our partnerships and working on sovereign deployments around the world: five of the six hyperscalers, the next phase of our NVIDIA partnership. We've talked about our OpenAI partnership, the recent Middle East announcements that we made.
We think we have a real good handle on what's going on in this space and how we can work together to help you be successful as you move forward. We all built the network when we viewed it as being critical: powering the internet. If you don't think the network's critical, that's not even the way to say it. We all know the network's critical because when you do have an outage, you get a few phone calls. No one says, "Oh, well, I'll just go home till tomorrow." It's so critical. The same way the critical infrastructure that we built for the internet, the critical infrastructure that we collectively built that powers the infrastructure in your environment, we're going to do the same thing for agentic AI.
Every one of you, when you think about what you need to actually be successful in this new phase, what you've told us in our discussions is you need to modernize your infrastructure, you need a new approach to safety and security, and you need more skills. We want to help you with all three of those as we go forward. As we think about how that comes to life, it comes to life through these outcomes that help you achieve this. We want to help you do three things. We want to help you build AI-ready data centers. We want to help you future-proof your workplace. We want to do that with an underlying layer of digital resilience. I'm going to quickly touch on all of these, and then Jeetu is going to go deep in each of these areas in just a few minutes.
When we say AI-ready data centers, we know that we have to transform our data centers to power AI workloads anywhere. We need seamless operations, observability, security everywhere. We want to help you do this with technologies like Secure AI Factory that we've announced with NVIDIA. Today, you're going to hear more about reimagining the data center with solutions like expanded AI pods, which we announced recently, Unified Nexus Dashboard. You're going to hear a lot about how we're going to help you build AI-ready data centers. Future-proof workplaces means we want to modernize everywhere people work and the technology they use. I think most of us know that when we came out of the pandemic, this approach where everybody worked remotely, there's variations to that now. There's some companies that are saying you have to come back. There's some companies saying you come back three days a week.
We know that the employee experience matters. It is so important. Whether it's secure campus and branch networking, smart buildings, collaboration devices, user protection, innovations like what we brought out with Wi-Fi 7 earlier this year, or the Nexus 9300 smart switches with integrated GPUs, today we're going to build on that momentum and talk to you about the biggest leap in our core networking in years across the campus, the branch, and our industrial portfolio. Then digital resilience. A year ago, we came together for the first time at Cisco Live and talked about Splunk and the integration of Splunk into Cisco and how that was going to be a baseline for delivering digital resilience with a lot of our other technologies like ThousandEyes and the integration of Splunk and Cisco to unify the data across your entire footprint.
Think about how AI and these models short-circuit all the problems that we had doing this in the past. Everybody was trying to build some big data lake with databases and with AI and with agentic and with some of the new protocols that we can use to access legacy data sources. We can do this in ways much faster than we ever could to help you keep your organization up and running. It includes assurance, observability, SecOps. A few weeks ago at RSA, we announced how we were going to bring Cisco XDR and Splunk even more closely together. We announced some agentic AI applications in security for threat resolution.
As we do this, we think we're going to give you a platform advantage because at some point, and Jeetu talks about this a lot, you can't just scale more people to fight the threats that we're going to be fighting on a global basis. We have to scale the machines to fight those threats. We're going to compound value that we can create for you, and you're going to hear more about that from Jeetu this week. As I think about what we can do together, I think it's important for us to first realize that we're all having to evolve at an incredible pace. I mean, this is changing so quickly. Between the technology transitions and the global geopolitical landscape, all of our lives are so much more complicated than they used to be.
As you try to achieve these outcomes and build the infrastructure to be ready for this AI era, the agentic AI era, quantum, whatever's next, you have to have security. You have to build it with integrity, and you have to build it with a consideration for humanity. That is underpinned by trust. We want to partner with you to bring that vision to life in a way that you feel good about. As I walk off today, I want to just reiterate two things. We believe that the era of agentic AI is only going to be secured by applying security services in the network to those workflows.
If you believe that, and Cisco is the only company that can fuse networking and security together seamlessly with technologies that we own, then hopefully you'll allow us to partner with you so that we truly can succeed together. Thank you again for being here with us this week.
The future with AI won't be a straight path. It'll twist, it'll turn, it'll zig, then zag. The answer isn't to avoid going forward. It's knowing how to navigate your way with certainty and speed, even when the path is complex. Cisco is powering the AI era by empowering you to reimagine data centers and workplaces with unrivaled infrastructure, security, and resilience, clearing your path to wide open possibility. Cisco, making AI work for you.
Please welcome President and Chief Product Officer Jeetu Patel. Good morning. How's everyone doing?
Great. Welcome to San Diego. Welcome to Cisco Live.
This is a particularly special one for me. Chuck mentioned this is one of the most consequential Cisco Lives we've had, but this one is extra special. Ask me why. My 14-year-old daughter is joining us here today for the very first time. You know I asked her, I'm like, "Hey, this is a really special day for Daddy. Can you make sure that you come to my event?" The first question she asked me is, "How long is it?" This morning she said, "Dad, it's 8:20 in the morning." Jokes apart, honey, I love you. Thank you for being here. It's very, very special. Like Carrie mentioned, stick around towards the end. We've got a really, really special guest, Kevin Weil is here. He's a good friend of mine.
He just recently joined the board, and I can't wait to share the perspectives that he has with you on what's happening with AI as well. Before that, we've got a very, very, very exciting set of innovations that the teams have been working extremely hard on that I want to make sure that we actually share with you. Let's start by thinking about what's happening in the world at a macro level because the next era of AI is here. If you think about what's happening today, we're moving from two and a half years ago when we had that seminal moment in November of 2020 when ChatGPT arrived. We had this kind of notion of chatbots that intelligently answered our questions. It felt like magic.
We are moving from that era to now where all of us are used to that, to an era where agents are autonomously going to be able to conduct tasks and jobs on behalf of humans. This is going to have a profound set of implications in the world that all of us will actually get affected by, and all of us will have an opportunity to really make a tremendous amount of positive light with. One of the big areas, though, that is going to be fundamentally impacted by this is the workforce. Today, 100% of the workforce that we actually participate in is humans. What we are going to have is a tremendous amount of augmentation to this workforce with agents, with applications, with robots, with humanoids that will actually now participate along with the workforce that we have.
What it does is it actually massively compounds the capacity of throughput that we can have. Eight billion humans will feel like we have the throughput capacity of 80 billion humans. What this is going to have is a fundamental set of implications on the underlying infrastructure that's going to be needed to power this movement. It's essentially going to have massive shifts in architectures. Like Chuck mentioned, there's going to be a shift in the architecture where it's not just going to be around public clouds. You'll actually have a good combination of public cloud and private cloud because there's a reacceleration, especially with inferencing workloads, where organizations want to have their own private cloud data center capacity built out for AI. Number two, AI is fundamentally going to be network constrained.
We need to make sure that we actually don't have this be bandwidth constrained. It's also going to be compute constrained, and it's going to be power constrained. Three is safety and security is going to be profoundly important in this era because if we don't trust AI, we're not going to use AI. The question you might ask is, what is Cisco focused on as we go through this entire movement, and why is it that we're going through this entire movement? One of the big reasons that this is important is because if you start to look at the world, there's only going to be two kinds of companies in the world. There's going to be companies that are extremely dexterous with the use of AI, and then there are going to be companies that really struggle for relevance.
We want to make sure that we can actually help you be the company in the first category by providing all the necessary infrastructure, safety, and security. What we've done, and we've thought long and hard about this, is what should Cisco uniquely focus on that's going to help you take advantage of this movement? There are three particular areas. The first one, like Chuck mentioned, is we want to make sure that we can provide you a world-class set of infrastructure for connectivity as well as for safety and security for this AI-ready data center build-out that's happening. We'll talk a lot more about that. The second area, and that's where, by the way, all the digital workers are going to live.
The second area is making sure that we future-proof workplaces where all of us work, whether it be a campus, whether it be a branch, whether it be an office, whether it be a car, what have you. Those two will need to collaborate with one another. We need to make sure that we have an underlying substrate of global connectivity. Third is we have to make sure that this infrastructure is resilient so that if there is an outage, you're able to get back up as fast as possible. Those are the kind of three core areas that we're going to be focusing on. We have a ton of innovation in each one of those three areas that we'll share with you. All of this is going to be powered by a foundation of AI.
Now, the question that you might ask is, what is unique about Cisco as we do this? What specifically are we doing that only Cisco can bring to the table? There are three things that we can think about that are very unique and are structural advantages for Cisco. The first one is this notion of providing a platform advantage. What I mean by a platform advantage is we have a tremendous breadth of capabilities of technology within our portfolio. Unless they all work in harmony with one another, you do not really get the most amount of value from it.
If you can actually reduce the marginal cost of ingestion with every single new technology that you buy from Cisco, and every single technology that you buy doesn't just add value by itself, but it also actually compounds value on things that you already own, what we've done is created a very, very advantageous platform advantage for you. That's number one. Number two, like Chuck mentioned, is this notion of silicon. We build our own custom silicon ASICs for the network. Why is that important? First, it actually gives you diversity so that you have choice and you don't have vendor lock-in by just having a single provider for silicon. The other thing that's actually really important is the way in which we've built out our silicon, the architecture that we've used to build it out has programmability.
What that means is every single time that you want to support a new kind of use case, we do not have to go out and etch it into silicon and do a tape out of silicon every single time. What you can do is make sure that you have programmability so that you can program the silicon and the cycle times compress and how you might be able to have a new use case afforded to you by that same chip. That actually creates a huge amount of acceleration of innovation because in addition to having a full top-to-bottom integration from the silicon to the network, to the security, to the models, to the applications, what we are also able to do is make the silicon itself programmable.
This is one area that if you do catch Marin, he's probably got the best jacket in this audience of anyone and the best chip in his back pocket. Make sure you talk to him. The third one is we need to make sure that AI is not an afterthought, but a foundational principle of how we build products, a foundational principle of how we operate. Those are the three key areas of structural advantage. Let's go into the details of what we've innovated in the AI-ready data centers. As we think about data centers, we are experiencing one of the largest expansions of data centers in human history right now. It's a phenomenal exponential curve that's only getting faster every single day. The question you might ask is, why is that?
What is the change in this pattern of data center consumption that's happening? Because we've been, for the past couple of years, the data center build-outs have been very aggressive on the training front. We've had a lot of large training runs we're doing with data centers. What you're now starting to see is it's not just about training the models. It's the inferencing of models where there's a huge amount of demand for data centers that's starting to build up. Here's what's different about the agentic era with the inferencing workloads. In inferencing right now, what you see is if you have the chat era, the inferencing peaks tend to be sporadic on the left-hand side of the chart. The interaction model is very sporadic. I ask a question, it gives me an answer. Every once in a while, you'll have peaks of compute.
In the agentic era, where you have agents conducting these jobs for you proactively, what you'll have is a sustained perpetual demand for inferencing capacity. When you have that kind of persistent inferencing need, you actually have to make sure that you change your entire assumptions on the way that the architecture has to get built out. There is a whole new level of demand signal for inferencing that's going to be needed for agentic AI, which is why what you're starting to see is there's build-outs that are happening all throughout the world. Now, Cisco happens to be foundational in these world's large data center build-outs. We've been helping hyperscalers for the past many years in building out data center capacity and making sure that we have connectivity intra-cluster for training runs that we can provide.
We've also been helping out NeoCloud providers who are specialists in building out data centers for AI workloads. Some of them are here in the room. We have a long-standing relationship with service providers. The beauty is we learn from all of these, and the enterprise also becomes a huge benefactor of it because all the learnings we get from hyperscalers and NeoClouds and service providers at scale can then be applied to the enterprise. A couple of weeks ago, I was out in the Middle East with Chuck, and we actually had some very strategic partnerships that we established with the government of Saudi Arabia and Humein, G42, Stargate, GOAE. This is going to continue to keep happening, and you should expect us to continue to participate in these massive build-outs of data centers because they're going to be network constrained.
They're going to have a prerequisite of safety and security, and we want to make sure that we can participate in that. Over the past six months, what's actually been staggering is the amount of innovation that we've actually had happen in this area of data centers. If you just look at the momentum we've had in the past six months, all of these capabilities that we have on the screen right now are capabilities that we just launched and are available today in the past six months. Now, my daughter Iman was very clear about, "Dad, don't go over." I am not going to go and drain this slide and go through every single one of these capabilities. What we'll do is we'll talk about just a few of them in a little bit more detail.
If you think about some of the really consequential capabilities that we've launched, the first one is providing you flexible capacity in the servers with GPUs. You can actually start small and scale fast based on your needs. You can start with two GPUs in your server, then go to four, then go to eight, and you can actually continue to keep scaling. The second area is hyperfabric, where we're fundamentally reimagining the data center lifecycle. The third is Intersight, which is one of the world's best management planes for managing our UCS servers and our compute business. That actually integrates with the rest of the platform.
The fourth is a completely revolutionary product that we built around safety and security for securing AI itself, which is a common substrate of security that you're going to have across every model, every agent, every cloud, every application that you might build with AI. Lastly, we wanted to make sure that we also provided a full-stack solution in a reference architecture with these AI pods. You can have compute, security, and storage and networking all provided in a single box. We'll partner with partners like Vast on the storage side and others, and we'll also make sure that we'll have compute, networking, and security in that pod. Those are some of the things. By the way, all of these are available today. You can use them. There you go. Thank you. One person's really excited. Chuck says there are three, actually, not one.
The question that a lot of you are here for is not just what we've delivered over the past six months. What are we planning to deliver as we move forward? Before we do that, it's not just about the enterprise. What we've also done is done an equal amount of momentum with service providers because service providers are the backbone of global connectivity. We've been partnered with service providers for decades now. In fact, we've been with them in this journey all along. We started with helping them with the voice transition, then data, then video, and now we're helping service providers with the move to AI and having them be able to assist their customers with this transition to AI.
In February, we had our Cisco Live in Amsterdam, and we announced a brand new architecture called the Agile Services Network, which essentially was infrastructure and validated designs for service providers. What we did over there was we not only provided this architecture, but we launched a couple of new Cisco 8000 Series routers at Cisco Live EMEA, and we've got a couple more that are coming right now at this event. We have this notion of optics for the long haul and ultra long haul. You can now do data center interconnects over 3,000 kilometers. We've got AI automation so that you can manage your estate effectively, and you've got provider connectivity assurance. These are some examples of what we're doing with service providers. You've seen a fair amount of innovation we've done on the enterprise side.
You've seen a fair amount of innovation we're doing on service providers. The question you're asking, though, is what are we announcing today? One of the areas that a lot of you have given us a lot of feedback on is, please make it simple for me to manage the Cisco estate. I am so excited to announce my first announcement. It's always special. The first announcement's always a special one, which is a unification of the Nexus Dashboard where Nexus OS and ACI Fabrics come together. This is a really big deal. This is something you've been asking for for a long time. Excited to let you know it's going to be shipping in July. We are going to have one brain for all of our data center network fabrics.
Now, what's also important is that we actually have an ecosystem of partners that we're working with. We are not just working in isolation from everyone else. One of the most important partners in this industry that we have is NVIDIA. We've got a long-standing relationship with NVIDIA. In fact, six years ago, we decided to actually have an NVIDIA chipset in our collaboration devices, and that's fundamentally changed how AI can be brought at the device level. That's not only helped our applications, it's also helped third-party applications that the device supports, like Microsoft Teams, that can actually take advantage of all the exciting AI capabilities. We had started with that, but over the course of the past year, we've really doubled down on the partnership. Let's talk about a few of these data center innovations we've done with NVIDIA.
The first one is an announcement that we made not too long ago, which was this notion of a Cisco AI Secure AI Factory with NVIDIA. Essentially, it's a reference architecture where we happen to be the first non-NVIDIA silicon provider as part of the reference architecture. We want to make sure that in this data center build-out, we're working closely with NVIDIA, and we are part of their reference architecture. We wanted to make sure, and this is available today, by the way, we wanted to make sure that we continue this build-out. What we now have is the next level of this where our switches are now fully integrated into the NVIDIA Spectrum X architecture, and they deliver low-latency intra-cluster communication.
What that means is I can have the NVIDIA switch that's working with, I can have the Cisco switch that's working with the NVIDIA NICs, and that's actually available for you to see in a demo in the world of solutions. Make sure you check it out. Essentially, this is a very exciting kind of time because we are continuing—the clock speed with which we're making these innovations is very, very fast. We're also doing more with respect to AI safety and security with NVIDIA. This is a really big deal. We launched our product AI Defense that I mentioned earlier back in January. Now we're announcing that open models developed with NVIDIA's NEMO framework can now be validated and secured by Cisco AI Defense. That's a big deal.
Hopefully, what you see is we're working together to accelerate the adoption of AI because the way that adoption of AI is going to scale is you need to make sure that safety and security is established and people trust these systems because if they trust these systems, they're going to use these systems. It's a very counterintuitive thing because in the past, security was always at odds with productivity. Now what you're starting to see is security is a prerequisite for adoption of AI. It's a prerequisite for success because if people don't trust the system, they're not going to use it. I hope you see how committed Cisco and NVIDIA are to advancing this AI market with the adoption. Our focus doesn't stop here on security. What we want to do is fundamentally reimagine security in today's hyperdistributed world.
We want to make sure that we rethink how this architecture is going to operate. The team has been working really hard on this particular problem, and they've put together a video. They've put together a video. Let's take a look.
The world has changed. Users, devices, and AI-powered applications are everywhere. But security is not. It's still something that comes in a box. Until now. Introducing Cisco Hybrid Mesh Firewall, protecting workloads wherever they live. Now policy enforcement can be distributed in our price-performance-leading firewall appliances as well as in virtual firewalls with eBPF in modern operating systems for visibility into the inner workings of AI-native applications or with a hardware accelerator placed right in the server complex. Perhaps most interestingly, we've fused security right into the fabric of the network using hardware-accelerated smart switches.
All of this is unified with our AI-native security cloud control platform, delivering granular application-specific controls. If it has a vulnerability, we compensate for it. If an attacker is moving laterally, we see it and stop them. We feed all the data into Splunk. It's distributed, optimized enforcement with unified management. Here's the best part. You don't have to rip out your existing firewalls because with Cisco's Hybrid Mesh Firewall, not only can we push rules into Cisco devices, we can also push them into your existing third-party firewalls. It's security fused into the fabric of the network, providing protection for every workload, every connection, every time. Cisco Hybrid Mesh Firewall, it's security for the way the world is now.
It's pretty cool, right? Now, here's the big takeaway. Cisco has categorically the industry's most complete hybrid mesh firewall solution in the market, period, full stop.
I'm so proud of this team and what they've done. We wanted to make sure that we actually continued to keep innovating in all dimensions of this hybrid mesh firewall. One of those areas is our core firewall offerings as well. I am delighted to announce that we have the ultra high-end firewall, the Cisco Secure Firewall 6100 series that's going to be available in October. It's the highest performance density for data centers. This is a long time coming. That's not all. We're also going to announce a Cisco Firewall 200 series for branches, also available in October. That's going to have an advanced on-box threat inspection for branches. With these two additions, we've completed the refresh of our entire lineup. We've never completed the refresh. We're already working on the next version.
This is the entire lineup at this point from the 200 series to the 1200 to the 3000, 4000 to the 6000 series, and then also the public and private firewalls. All of those are refreshed. All of them have the industry's best price-performance ratio. We think you're going to love them. They're all managed by one management plane called Security Cloud Control, which is also the management plane that manages the rest of the hybrid mesh firewall. How cool is that? What we wanted to do was we wanted to make sure that some of these core technologies that we're building solve some really big problems that we're facing as a society in the market right now. One of the big problems that we're facing is it's really hard to patch networking gear. What we wanted to do was do something completely innovative.
As we had started building HyperShield, there was this kind of amazing capability that we had built that could allow you to provide shielding for a vulnerability that comes out. Here's a really interesting thing. When a vulnerability is exposed, it takes about 45 days to patch the vulnerability, but it only takes three days for an exploit. You have this time where you're just exposed when a vulnerability has actually been announced in the market. We wanted to do something completely revolutionary. I'm announcing LiveProtect, which essentially is a vulnerability shielding mechanism for Cisco networking devices. We're going to start this with Nexus OS, and then we're going to make sure that this will actually go to our entire estate over time.
What this means is when you have a vulnerability, there will be a compensating control, a shield that will be applied to the vulnerability within minutes. Then over time, as you have a patch, the vulnerability will get taken away. Amazing innovation, LiveProtect. It's going to be available for Nexus OS in September of 2025. If you notice a pattern, thank you. If you notice a pattern, I'm actually telling you about all of these announcements that are coming out, and they're all imminent. These are not very far out. It's one thing to hear from me. I get my paycheck from Cisco, but let's hear from a customer who's actually completely reimagining the data center with Cisco technology. Alan Rosa is the SVP and CISO of CVS Health. Alan, come on up on stage.
How are you, sir?
I'm good.
It's good to see you. Mesh Firewalls, Unified. So much Firewall. And LiveProtect? LiveProtect. That was not scripted. Thank you, sir. How has AI fundamentally changed your approach with modernizing your data center, Alan? Tell everyone about it.
You're doing some really cool stuff. We're at an interesting inflection point.
I like your kicks, by the way.
Oh, thank you.
They're very white.
We're at an interesting inflection point in that we have a chance to modernize our eight existing enterprise data centers, consolidate them, and move into new facilities. We're taking that opportunity to move beyond the traditional workload, embedding AI capabilities inside of our fabrics. We're thinking about GPUs first, and how do we prepare to deploy more advanced agentic solutions. On top of that, we're taking a security lens to it.
As we think about deploying a workload, how we evolve a workload, how we deliver it, it's all going to be done in a secure context. Our teams are very excited about that opportunity.
How is networking changing as it pertains to that as well?
Obviously, the modernization journey is there. High degrees of automation. Our teams are experimenting with agentic solutions now. Our goal is to make the network far more autonomous than it is today and then create opportunities for our technical operations teams not to compromise stability, resiliency, and security. Just last week, we announced a 10-year, $20 billion investment to advance AI and technology. I would say to channel my inner Chuck, we mean it. It's fantastic.
Talk to us. You're the CISO.
Talk to us about the security domain and how that plays into this whole AI arena.
We took a unique approach to the way we built our operating model. It starts with our people. We have some of the best engineers in the world, really thoughtful leaders that think about how we put this stuff, putting the customer, the patient, and the member first. When we combined our teams, we have our security and our privacy engine leaders, our infrastructure leaders, our digital workplace leaders, our enterprise architect teams together. These teams, they look at a problem by design. It's our security leaders influencing infrastructure designs. It's our infrastructure leaders influencing security decisions. It's our colleague, digital workplace leaders thinking about how is that going to represent itself, the people who consume our services. It allows us to build more fluid products.
We're seeing that pay off the last couple of ye ars.
This seems like a fair amount of innovation. What is the cultural implication that you're kind of how do you deal with the culture of change internally within your organization as you're thinking through this?
The thing to understand about CVS Health is that we are a people company. Folks come to us when they're facing obstacles in their health journey. These are critical moments for them, and they trust us to be there for them. The way we think about our members, our patients, and our customers is the same way we think about our colleagues. We want to create meaningful career journeys for them. They feel like they're owning their work. They're allowed to innovate and experiment.
We're putting the work in the hands of the people we trust the most, which is our CVS colleagues. There will always be room for providers and partners, but we're going to be far more strategic and targeted about who we invite to partner with us because our colleagues need to lead.
You know, I can answer this is off script, but I'm going to tell you a little bit of a story. My mother passed away like two and a half years ago. She was, of any brand in the world, she was the most brand loyal for CVS. One time, we had to change pharmacies for like a quick month, and she would not have it. She's like, "I need CVS. I need you to make sure that my medications are from CVS." The lady's really nice.
That translates to customer sentiment in a very, very good way. What is next? What is next with CVS?
We are going to continue, we are aspiring to build a vertically integrated digital health experience, which takes the best of our health benefits, our PBM, our pharmacies, bring them together so that it is easier for our members and our patients to engage with us. We are going to offer that out to the public. We are hoping that drives more visibility, more connection with these folks. On top of that, it is continue to evolve our internal working culture. I mean, as I have said a million times before, we are not going far without the best colleagues in the world. I think we have the right people working on this.
Alan, you are doing an amazing job. Thank you again. What an honor to serve you, my friend. Thank you. Take care.
All right, next, I want to talk to you about how we are future-proofing the workplace where all of us operate, whether it be a campus, branch, office, home office, your car, whatever place. If you think about the workplace for tomorrow, it's going to be very different from the one it is today because the workplace is going to have a fundamental augmentation of agents and robots and IoT devices. There are going to be different ways that people collaborate within the workplace. You're going to have constraints around power. You're going to have threats around how bad actors and adversaries attack your workplace. All of these pieces are going to be things that have a very different set of demands from your infrastructure. What we wanted to do was focus on three very specific things in future-proofing workplaces.
Number one, we want to make sure that we can provide operational simplicity powered by AI. This is something that all of you have asked for for a long time. We've got some really exciting things to dramatically simplify the way in which you work and the way in which you manage your estate and your infrastructure. Number two, we want to make sure that we have scalable devices that are ready for AI. We've got some really exciting new devices that we're announcing. Number three, like Chuck mentioned, we're fusing security into the fabric of the network, just like we did in data center. Let's start with the operational simplicity powered by AI is the first one out. This is an area which is near and dear to a lot of you. Basically, what we're doing is we're unifying Catalyst and Meraki.
I was thinking there would be a lot of excitement for that one at the hardware level with license management, with licensing as well as with management. Now, let's dig into the unified management because that's an area that a lot of you have had a lot of energy on. We are announcing for the very first time a unified cloud management with a single dashboard for Meraki, Catalyst, and all next-gen devices. We've been talking about this for years. What I had told the team was, I'm not going there and giving them another date. This is available now. There is more because we are also giving you assurance for every connected experience by announcing a multi-layered assurance capability.
I'm going to take a moment over here and explain what this means because when you think about assurance, we've been very good when you think about managing complexity when you have an experience that's not perfect and you want to pinpoint exactly why that experience is not perfect. We've been very good at managing and depicting what's happening in the unowned infrastructure. What we've not always been the best at is the owned infrastructure. What this announcement does is it expands our visibility for both owned and unowned for networks, for clouds, for wired and wireless devices. By the way, at this point in time, we are by far the most comprehensive when it comes to assurance being provided for owned and unowned devices. This is going to be available this month.
The reality is, even though this is the best end-to-end assurance and the best kind of converged management, we did not want to just simplify IT experiences for you. We want to fundamentally reimagine it. This is a moment that we have actually been very excited about and working hard on for at least a year and a half. It gives me so much pleasure and so much joy to introduce this concept of agentic ops. Basically, what agentic ops is, is the ability to have cross-domain visibility and management, this notion of multi-player infrastructure management. If you think about your day-to-day job, it is not like you are working in a silo by yourself. You work with many other people troubleshooting when something goes wrong. However, all the tools are single-player tools. We wanted to actually build a tool inherently from the ground up that was a multi-player collaborative tool.
That's going to be powered by a purpose-built AI model for networking. This is very, very cool. Let me get into the details of this because what we're announcing, first and foremost, is a purpose-built model for network experts called the Deep Network Model. Now, what this is, recently Sam Altman was at an event, and he had a very insightful line that he talked about, which is the future of models is going to be that the models are going to be very small in size, some of them. They might have a trillion token context window, and they'll be integrated into everything. Because today, if you think about it, a lot of the models act as databases. In the future, you want to make sure that the models can go smaller in certain use cases.
This is one example of one of those models, which is a deep network model. It's purpose-built for networking, for expert accuracy, and it actually performs 20% better than some of the other models as it pertains to the domain of networking. Now, this is great, but what does this do? What is this going to give us as an end outcome? To showcase this model, what we are announcing for the very first time is a completely reimagined way that AI is going to help you manage your entire estate, and it's called AI Canvas. It's a completely reimagined user interface. In fact, it is a generative UI. That means that it generates dashboards on the fly in a multimodal way.
Rather than me talking about it, what I thought I'd do is I'd have our head of AI and platforms, DJ, come up on stage and give you a live demo of this. DJ, come on up. All right, buddy. All right. What do you have to show us?
All right. Thanks for having me here, Jeetu. This is really exciting. I want to show you what the team's been working on. All right. Let's get into this. I'm going to hop in here. This is AI Canvas. What you're seeing here on the left-hand side is the AI assistant where you can use natural language to be able to communicate with the canvas. On the right is where the agents and humans are going to work together to be able to solve problems. All right.
Now, let's imagine that there is a network outage, and you want to be able to start troubleshooting this. Normally, what happens is you actually have a ServiceNow ticket that's raised, and the ticket has all of the details about what's going on. What I'm going to do is I'm going to post this message here and tell the assistant to go fetch that data. Automatically, you'll notice that it's gone out to ServiceNow and pulled all of that information right here. Folks, just so you know, this is a live demo. I'm keeping my fingers crossed as I get into this. Normally, when an admin sees this, they have to now go to different screens to be able to go out and start troubleshooting. With AI Canvas, we're making this incredibly simple.
I'm just going to type in here, "Troubleshoot this ticket." Now, this is where it starts to get really cool. The deep network model that you just announced is going to start breaking down this ticket and analyzing that it needs more data. It now is going to go to Meraki and pick out the packet loss trends, and it's going to generate the UI for you. You just saw that it created a widget. That widget was not something you built. It actually generated the widget. AI generates that widget, recognizes that the next thing that you're going to need is a time series data. You can see it's constantly thinking right here at the bottom, and it's pulling data that it needs.
It recognizes now that it needs data from ThousandEyes, not just from Meraki, to be able to show you exactly a path visualization that tells you where this outage might be happening. As soon as you see that, you can see, boom, it's going to generate that path visualization for you and lays it all out. Now, here's the other part. The assistant's also going to start giving you remediation suggestions of what you need to do to be able to start debugging this issue. You also mentioned that troubleshooting almost always is a multiplayer sport. If you wanted to invite other team members to jump in over here, it's really simple. You invite them to the board. I'm going to invite Lizzie here. With a single click, it's going to generate a summary of all things that I have done so far.
I don't have to sit down and type out an email to get Lizzie on the same page. With a single click, Lizzie can jump on the board and start working together with me.
Completely collaborative.
Completely collaborative, completely multip layer.
That's very cool. All right.
That's pretty cool, right? Okay. Here's the thing. Lizzie might come in and say, "Listen, I like how you're doing this, DJ, but you know what? I'm not so sure. I want to verify using another source of data." She might come in here and say, "I want to correlate this with the application logs to see what's happening." Now, these logs are not inside of ThousandEyes or Meraki, but these logs are inside of, say, for instance, here, Splunk.
What the deep network model is going to do, it's going to recognize that this data is not here and it needs to go to a different place, and it's going to pull that information and start to render that right here. It's going to happen any moment now. There it is. All right. It's really simple. Here, you could go ahead, drag and drop this into your canvas, and what you're seeing here is the performance of that application is starting to drop. That's your red line. The blue line is indicating your packet loss percentage that's starting to go up. Clearly, there's a correlation. Yeah. Exactly. The interesting part about AI Canvas here is it's not just for monitoring. You can start to take actions just using the same console as well. This is where the agentic ops kicks in.
Let me show you how. Now, back here, the assistant said, "Hey, you can go ahead, take these remediation actions." It says, "You got to configure a QoS policy on the MX device." I'm going to go ahead and ask the assistant to apply the QoS policy, or the QoS stands for quality of service. Again, it's going to go out, identify exactly which device it needs to apply this to, or a set of devices, and it's going to automatically apply that and give me a widget that I can now go ahead, add back to my canvas. Now, there's something really cool over here. You'll notice that it actually gives you an opportunity to revert a change. This is where we're making sure that there's always a human in the loop that can come in and change things up based on what they're seeing.
Now, back to the dashboard over here. Now that we've applied the QoS policy and the agents are going out and doing the job for you, you want to make sure that these things are happening the way you expect it to. I am going to come back in here and say, "Hey, I want to keep track of this. I want to monitor how this change is affecting my environment." This is where these things are not just a one-time widget. The agentic ops ensures that it's constantly going back out, fetching you live refreshed data so it's always current whenever you come into this board. You are going to see it's thinking right there, and it's going to pop back out in a second.
You're going to see that the graphs, the data from the Splunk and the application logic is going to go through a change. Boom. You notice that the packet loss has dropped.
Amazing.
The application's back up and running. All of this was live, and finally, you have to write a report. Thank you. With a single click of a button, you're going to be able to generate a report, and you can add that report. Again, you don't have to sit down and write this. It takes a lot of time for the admins to do this. It's going to automatically generate that, and you can download that as a PDF, and you're off to the races.
Can I tell you a story?
When he said to me, he's like, "Gito, I want to do this demo live." I'm like, "Are you sure?" Because it's like 20,000 people connecting to the same network. It's going to be embarrassing if it doesn't work. He's like, "I don't think it's credible if we don't do it live. We have to do a live demo." DJ, sump up. Hang on, hang on. You're not going anywhere. When are you going to ship this?
This is going to be in the hands of customers by October of this year, right around the corner.
That's like, you know, you're the longest time frame of anyone. Can you pull it up a little bit? We got some people who are--do you guys want it earlier? Yes. All right, all right. You'll take it in. Okay.
Here's one thing, though.
The deep network model is going to be available by the end of this month for all CCIE users at the Cisco U portal. All right.
He redeemed himself. Let's give it up for DJ Sampath , folks.
Thank you so much.
Thanks, DJ. Thank you, buddy. Great job. If there's one thing you take away from this, it's that with agentic ops, the way in which you run your network will never be the same again. This is so cool because we're just getting warmed up. We're just getting started on this area. You're going to see a lot of innovation. The beauty about it is it keeps getting better and better the more you use it. Okay. Now let's talk about scalable network devices that are ready for AI. We talked about operational simplicity. Let's talk about scalable network devices.
Today, we're introducing one of the largest refreshes of networking devices in Cisco's history. Everything from campus, branch, and industrial IoT, we've got some refreshes there for you. Rather than me talk about it, let's actually play the video that the team's put together for this. They were so excited to talk about every single one of the innovations that they've been working so hard on. Let's take a look.
Introducing a new lineup for the AI era. Smart switches. To scale networks simply and securely. Secure routers to protect data and devices across thousands of sites. Intelligent Wi-Fi 7 to delight even the largest crowds. Cloud-native campus gateways for seamless roaming in the campus. Wi-Fi and URWB for ultra-reliable zero-loss access everywhere. Advanced firewalls for distributed threat detection. New rugged switches to power AI-ready factories. The secure network for AI-ready campus, branch, and industrial IoT.
All of these devices orderable this month. It's amazing. It's amazing. Congratulations to the team. We hope you love them as much as we've enjoyed building them. Now, what I want to do is I want to actually dig into a little bit of detail on one device, which is a whole brand new category of a device, which is the first one, smart switches. Essentially, what smart switches are, they're built on Silicon One, of course, but they actually have an isolated compute space which can run HyperShield over time. You now have a switch that has isolated compute that can run other workloads like security. That is completely revolutionary. That's not something that anyone else is doing in the market. We're the first ones to do it. We did it in the data center, and now we're bringing it to campus and branch.
Super excited. We hope you order a bunch of these. All right. Next, let's talk about this is, by the way, this will fundamentally kind of change the way you future-proof your network. As you think about the workplace, it's much more than the network. Experiences that people have in the workplace really matter. One of the things that we've developed is a family of AI-powered collaboration devices, like I mentioned. These collaboration devices essentially have NVIDIA chipsets in them. They've got AI capabilities in them. They're the most beautiful, gorgeous devices. I don't even know what I would do if I didn't have one of them because when you're on calls for long amounts of time, the fatigue factor goes down. The beauty is these are now completely integrated into an open ecosystem. Microsoft Team Rooms runs natively on these devices.
In fact, so many organizations that have standardized on Microsoft Team Rooms have actually standardized on our devices. It's hundreds of millions of dollars of value that's been created within a very short amount of time. The real power is when these are combined with our networking capability. What we've done is we've actually made sure that things such as switches and cameras get augmented to this portfolio, and these all become sensors on the network. What I mean by sensors on the network is they provide you important data about your facility, about your campus, things like temperature, things like call quality, things like occupancy data. That data actually allows you to make sure that your environment, your workplace can be more and more effective and can be managed in an easier way.
Now, one of the things that we had really focused on about a year ago was this concept of distance zero. Distance zero means when there are two people on a call, we should make sure that the distance between those two people goes away, that they feel like they're right there with each other. What we want to do is we want to make sure that the distance zero, which is an end-user-focused concept, we also bring for the IT administrator. We wanted to have one-click distance zero for the IT administrator. We did this last year with the microphone where we actually had with one click, you can just easily configure this. Just plug it into a switch and it auto-configures itself. You don't have to do anything. It's all AI-configured.
We started that with the microphones, and we've got another exciting announcement for you. Let's roll video. Let me tell you what that is because that is a bunch of cameras that auto-configure themselves to provide you cinematic-quality meetings. What does a cinematic meeting mean? Have you ever noticed that it's much more tiring to be in a meeting for an hour than watch a movie for an hour? Why is that? Because in a movie, your frames keep changing, and it actually resets your brain every so often. In a video call, you're just looking at the same angle for the full hour, so it gets fatiguing. What this does is if you have a conference room with multiple people, all these cameras auto-configure, and they work with each other, and they give you a cinematic quality to the meeting.
If someone is speaking, they'll give you a close-up view. If someone else is speaking, it's going to give you a zoomed-out view, and it's automatically going to do that. It's completely magical. Cinematic meetings with pan, tilt, and zoom cameras, PTZ. And it's one of the best devices in the world. Folks, I think that you are really defying logic if you ignore Cisco devices in your estate in the workplace. It makes no sense. You should just make sure that you use Cisco devices in the workplace. Really. I'm telling you. All right. Now let's take a look at how we fuse security in the workplace because not only are these the best devices and the best networks, but they're also the most secure. How do you fuse security into the workplace?
One of the areas that the industry has focused on quite a bit is this notion of zero trust, which is least privileged access. Only give people access to the things that they need access to and no more so that they can get their job done. We started zero trust network access with this notion of a user connecting to an application should only be given the appropriate amount of permissioning and no more so that they can get the job done. Zero trust for a user connecting to an application. Now we want to expand that aperture a little bit because we want to reimagine zero trust not just for users. We want to imagine it for things, and we want to imagine it for agents. We want to make sure that that's actually securely done.
We have a great video on this one as well. Let's take a look.
Today, zero trust is the new normal. You get the least access needed to get the job done, nothing more. It sounds easy, but it's not. Users and things are everywhere, accessing sensitive assets on any device from anywhere. But so are attackers logging in with your credentials right now. We took a different approach with Cisco Universal ZTNA. Whether you are a developer on a managed or unmanaged device, or an agentic digital worker approving a large purchase order, or an in-office printer, whether you're remote or in the office, you get the same policy and controls for over 1,200 Gen AI applications. But can you enforce zero trust for something that doesn't have an identity? We can.
With ICE Tags from the network, we always know who's who, even if the who is a printer. A single lightweight client provides multiple functions to deliver a seamless experience to any application, all with the magic of Duo built right in. We do not just say we deliver a great experience. We measure it with ThousandEyes. Even if it's not great, we show you why. With Cisco Identity Intelligence, we make sure you are who you say you are every time. Zero friction, zero imposters, zero downtime. Cisco Universal Zero Trust Network Access. Always aware, barely there.
Pretty cool, huh? If there's one thing with all these security innovations you just saw, if there's one thing you take away, it's Cisco is building the most comprehensive security platform for agentic AI. There's no one else who even comes close.
We are going to continue to keep innovating at the velocity, and that velocity and the rate of change is going to keep getting increased. The team is hungry to make sure that we serve you right, and the team is hungry to make sure that you are secure in everything that you do. Next, let's talk about digital resilience, which is the third of the three key outcomes and priorities that we want to focus on. Now, if you think about when you have an outage within your organization, the first few hours of the outage sometimes can be spent just determining what the challenge is. Is the challenge the fact that you've actually got something wrong with your network? Do you have a security breach? Is there an API overage? Or is there a bug in the application?
The amount of time it takes to detect these outages is very long, and the complexity that you have, you all manage a very complex environment. The complexity actually creates a lot of risk for business continuity. This notion of problem diagnosis is really hard. The data, which used to be largely in data centers, is now also going to be in a bunch of other places. Traditionally, data volume was constrained to data centers, but now you have data on the edge. With agentic AI, you've got an explosion of data everywhere. Essentially, when you think about this with adding of robotics and with making sure that you have more industrial IoT in your factory floors, digital resilience and making sure that your estate is resilient is essentially a data problem.
One of the big reasons that we acquired Splunk for a small price of $28 billion was to make sure that we could actually take all this telemetry and data across multiple domains and be able to correlate it. If you have network data and infrastructure data and application data, but also have user and device data and operations data for IT operations and security data, let's take all of that data, keep it in place, and correlate it. When you think about doing this, here's what you can do with all that data. The first thing you have to do when you have all this data is you have to distill it down because the volume of this data is so high that if you actually have a very large corpus of data, it's very hard to get the signal from the noise.
You have to distill it down. Low volume, high fidelity data is the first step. The second step is you have to make sure that you can correlate that data across your data while keeping the data in place. The third thing is you have to unleash AI on that data. When you do that, there are a few really exciting outcomes that start to emerge. You get assurance across everything. You get observability everywhere. The security operations center can now get fully reimagined. What Cisco and Splunk have been working very hard on in the past year that we've actually made this acquisition is to ensure that not only do we not slow down the innovation velocity of Splunk, but we make sure that we actually have 1 + 1 = 11.
Just in the past few months, what we've done is we've actually had a ton of innovation. Telemetry from different sources that can be pulled together in Splunk make sure that you can have correlation across that telemetry. We did not want to stop there. Today, what we're doing is we're announcing a brand new integration for Splunk Observability with Catalyst, Meraki, and ThousandEyes. You'll be able to now correlate network data and network health with IT service and business health data. ThousandEyes is available now, Catalyst and Meraki available in September of 2025. It's very cool. This next one you're going to like even more. One of the things that you typically tend to use Splunk for is you ingest logs from firewalls.
You just saw that we got a completely refreshed lineup of firewalls, and we've got the best hybrid mesh firewall. Wouldn't it be great if we made it more economical for you to ingest Cisco firewall logs? For that, what we're announcing is that if your firewall log that you're trying to ingest into Splunk is from Cisco, it's going to be for free. People like free. This is going to be available in August. As people start using Splunk more, one of the things that you find is they aren't just using it in the cloud. Sometimes people want to use Splunk on-prem. We said, what can we do to simplify the use of Splunk even on-prem? We are announcing for the very first time a Splunk Pod, which is a validated design and simplified ordering for Splunk platform with Cisco Compute.
You will basically have a validated design for a Splunk Pod on UCS servers available in June of 2025, this month. This was, by the way, last year when I was at Cisco Live. This was a huge request by a lot of the Splunk customers. Now, just like we had security for AI, we want to make sure that these AI agents that get built out, there's going to be a fair amount of visibility that's going to be needed for these agents. We're announcing observability for AI. It's going to be a comprehensive visibility into the AI estate. This is going to be available in September of 2025. What I mean by this is you will have visibility into your GPUs and the utilization of GPUs, your models and how they're being utilized.
You'll have full visibility into the entire estate for AI. That is going to be available. Very excited about that. Now let's switch from observability to security operations. Folks, we must fundamentally reimagine the operations of security. We have to do it, like Chuck mentioned, at machine scale because the attacks are coming at us at machine scale. You can't go out and defend at human scale. It just won't work. What we've done is we are trying to make sure that we can completely reimagine the security operations center. The way that we're going to do this is we're going to be delivering our unified threat detection and investigation and response platforms. Essentially, we're powering the SOC of the future with a market-leading SIM and a completely revolutionary XDR put together.
Now, the amazing part about this is when you start thinking about the SOC of the future, you have to think about the model underlying it and how it's going to power the SOC in this age of AI. We are taking it further because we want to make sure that we have agentic AI in the SOC, and we want to make sure that these models are bespoke. Back at RSA, a couple of months ago, we announced with our foundation AI research team a security model. The best part about this model is we open-sourced it. In the past couple of months, we've had over 40,000 downloads on Hugging Face on this model. It's purpose-built for security. It's easily customizable, and it's highly efficient. We trained it. We pre-trained it on an 8 billion parameter model, which is pretty small.
We actually took a corpus of data of 900 billion tokens, but only gave it the relevant 5 billion tokens we trained it on. It actually performed better or as well as a 70 billion parameter model because you've got only security problems that you're trying to train it on. It actually was so amazing because it ran on a single A100 GPU. Think about the cost curve going down of running these models when they're small and bespoke and specialized. We said we want to do a little bit more with that. We quantized that model. Instead of being 16-bit, we actually quantized it to 8 bits and 4 bits, which means you can be four times as efficient. This model now, folks, you can run it on a single laptop with a CPU. You don't even need a GPU to run the model.
It's amazing what it can do. Essentially, that'll actually be the core foundation on top of which you can start thinking about powering the SOC. If you are interested in playing with the model, go to Hugging Face. It's called FoundationSec. The AI research team, that was a team that actually was acquired from an acquisition we made called Robust Intelligence, is doing a fantastic job. Take a look at it. Basically, what have you heard so far in the past 58 minutes? What you've heard is we have AI-ready data centers, and we've got amazing innovation over there. We've got amazing innovation for future-proofing workplaces. We've done a lot of work with service providers to provide the secure global connectivity, and we've got resilience. We've got digital resilience that you can make sure that we deliver.
Now, one of the areas that you folks had given us a lot of feedback on that hopefully you saw some payoff today on that we are executing is make management simplified for me. What did we do? We took Nexus OS and ACI, and we merged them together. We took Catalyst and Meraki, and we merged those together so that you have one common way of managing it. We used to have a year ago about 15 different ways that you could manage your security. Now you've got security cloud control. We think we can do even better. I've got a surprise for you. This is the last one. We're announcing Cisco Cloud Control. Cisco Cloud Control is unified identity, common design language, cross-product workflows, agentic ops, all in one management plane. This is something, and this is what it looks like.
It's essentially something that is going to completely simplify management for you. The core kernel of it will be AI Canvas, which is what you saw from DJ. What do you think? Hopefully, what you've seen over here, if there's one thing you take away from the keynote is if you think about the entire AI stack, you've got silicon, you've got the network on top of that, you've got compute, you've got data, you've got models, you've got applications that are being built. They have to be with the right level of visibility and observability, and they have to have the right level of safety and security. We play in every single one of these areas.
We are spending billions of dollars a year on the stack so that you can stay safe, you can stay innovative, and you can actually be dexterous in your use with AI so that you're ahead of the market in what you do. That is the innovation payload that the team has. Now, I have one request for you. This doesn't come easily. It takes a lot of kind of sweat and hard work from the team. Unfortunately, because these rooms can only hold X number of people, and we try to spend our budget on building products rather than travel, I want to make sure that the engineering and the product and the design teams can see if you folks feel like this was a fun innovation, can I actually have the lights turned on and you folks give them a big round of applause?
They're going to love it. They're going to love it. This is what we live for. We love to make sure the customers are delighted with our products. With that, I've saved the best for last. My good friend Kevin, Chief Product Officer of OpenAI, let's make sure we welcome him on stage. Hey, welcome to an enterprise conference, man.
Thank you for having me. This is awesome. By the way, I really like the AI Canvas. I mean, the keynote in general is fantastic, but the AI Canvas, also DJ and the bravery to come up and do a live video, our live demo. We livestream all of our launches. It's slightly terrifying. I've got a lot of respect for it, and that really looked great.
We are actually learning from you folks that we're going to livestream one of the AI Canvas launches as well. We've done this with the product team. We tried to copy you blatantly. We figured it'd be okay.
We started just to say we kind of did this thing around Christmas. We had all these launches bunched together, and we were like, what are we going to do? How are we going to get all of this out? We said, okay, this is crazy, but what if we just did 12 days of launches in a row, and every day we livestream? It just about killed me. It turned out we learned something, which was livestreams actually are a really fantastic way to launch a product because you get to communicate kind of the whole product end to end. You get to show what's going on. It's live, so it's very tangible and real. It's actually turned out to be a really good format for us launching new things.
No, I think one of the things we struggle with, frankly, is the innovation velocity is so high, and I don't think we've cracked the code yet of getting a million customers, which is what we have, to really get up to speed on our products on a weekly basis. We haven't cracked the code yet, but livestream might be the thing that actually makes that happen.
Yeah, much more entertaining than a blog post, right?
Much more entertaining. Hey, you and I have known each other for a long time. Thank you for being here at the event. Let me just brag about Kevin. Kevin used to be the head of product at Instagram, head of product at Twitter. He was the president of Planet and so many other amazing things he's done. He goes to OpenAI a year ago. Tell us about your journey and why you chose to go to OpenAI from all these other places you were at. How has it been in the past year?
It's been fantastic. I mean, I've been really lucky in my career to get to work at a lot of innovative, very mission-driven places. It's the thing I look for. You want to be a part of something bigger than you in the world and leave the place better than you found it, contribute. I wasn't looking for it. I was actually trying to take a little bit of time off. I'd known Sam for a little while, and we talked on the phone. I would usually call him actually when I was thinking about doing something new because he's such a great thinker about the future. This time when we talked, he was like, actually. As I came in and met a couple of people, I came back to him and was just like, there's no commitment yet.
I haven't even really interviewed. If you give me this job, I'm just going to say yes. At the risk of negotiating against myself here, I'm just going to say yes because this is the coolest, most impactful thing I could imagine doing. It really has been. I mean, I've learned so much. It's an absolutely incredible company. We're all seeing AI change the way that we live, that we operate in our personal lives, that every business operates. Just out of curiosity, who uses ChatGPT? That's awesome.
Actually, we can't see, so can you just, with a round of applause, who uses ChatGPT? Okay, great. Yeah, for a product that's.
I know my daughter uses it.
Two, two and a half years old to have over 500 million people using it every week. It's just, it's a privilege. It's a responsibility we take really seriously. As a product builder, it's an incredible amount of fun.
Yeah, you folks have done such an amazing job with it. You're an inspiration to the industry in how well you're building the products and the speed at which you're building them. There are constraints that you folks have. I know that there are times when you folks are running around trying to find GPUs even at your scale because.
Sam will occasionally post on Twitter that we're looking for more GPUs. Yeah.
There is clearly a constraint on infrastructure. If you did not have a constraint on infrastructure, what would this look like? Is that your biggest kind of impediment right now for really blowing this up?
Yeah, I mean, it is. I think one of the things that you see us doing is big projects like Stargate, which we announced this Stargate project with President Trump and Masa and Larry Ellison and other folks in the.
We were there.
You were there, absolutely. We are together putting an investment of something like $500 billion in compute, data centers, infrastructure because it is so critical to the future of AI. Frankly, we see over and over again, the lesson that we have learned is the more compute we have, the more powerful the models that we can build, the more powerful the models we can build, the more value we can provide to every person and every business. We have more or less an insatiable appetite for compute. It is not just the models because that is the training side of it. It is also using the models in the applications that we build in ChatGPT and the other products because we are a funny company in that most companies that have a product that has a paywall, over time they move things behind the paywall. We do the opposite.
Things start behind the paywall often because they use a lot of compute, and we need them to start in a more limited way. As we have more compute, we make them free. With more compute, we're able to offer more things to all of the people that use ChatGPT. I'll give you one more thing, which is today, and we can talk about reasoning models and some of the big advances in the models themselves, but today, the models, we went from GPT-4, which you ask it a question and system one thinking, and you get an answer. The reasoning models that we have are a little more system two. You ask them questions, and they can stop and reason and build hypotheses, validate or refute those hypotheses, and they'll think. They can use lots of tools.
Still, they're thinking for like a minute, and then they'll give you a response. We have a product called Deep Research. It thinks for maybe 20 minutes to give you like a 20-page written report on any complex topic that you want to know about.
By the way, I think I might be the consumer of 40% of the compute research.
Amazing. Imagine you give these models not just a second to think or a minute to think, but imagine you give them an hour, a day. Imagine you had a model that could work for a month or a year just on some super important, really hard problem. I think we're going to make amazing advances in really critical things like fundamental science and biology with models that can think that long. It takes a huge amount of compute to do it. My answer is with more compute, I mean, we could do so many more things. That is why we're investing, and we're investing together.
Right. How constrained is this from a network infrastructure? What do you think of safety and security in the whole?
I mean, all of this is critical, right? It can't just be GPUs. You've got to have super robust, highly performant networking infrastructure. You've got to have security in everything that you do because at the end of the day, we produce a set of weights, and people are trusting us with really important information. It's got to be the whole package, right? It's not just one thing. All of it has to come together so that we can build applications that people trust, and you all can build businesses that can rely on it.
Talk a little bit about you've now been there for a year. What's your favorite product launch that you've done? What's been the thing that's most exciting to you? The next question I want to ask that you can think about is what's the most surprising thing that you didn't think you would see at OpenAI that you ended up seeing?
Ooh, that's a good question. I think probably my favorite product that we've launched so far, and I hope I'm giving you a different answer in a little while, but is this Deep Research product. Because it's the first example. People are talking about agents, and it's the first example of an agent that really qualitatively changed something for me. I use ChatGPT all the time, as I hope all of you do as well. It saves me a bunch of time here and there. It's like helping me brainstorm, write emails, look things up, learn about new topics. It's probably saving me two minutes, five minutes, ten minutes here and there. This Deep Research product that we launched allows you to ask the model any arbitrary tough question.
Give me an analysis across all 50 states of the tax differences in this one particular thing if I'm going to go into this market and break it all down on a table and also talk to me about break it down in these other three ways. Stuff that you would have to go if you were going to go do this yourself or you were going to assign an intern to go do it would take a week. You'd be looking at hundreds of different sites and cobbling together information. Actually, that's what the model goes and does. We built a model that's really good at doing internet research. It can go and look at hundreds of different sites, figure out what it's learned, how well that answers your question based on what it's got.
Maybe it realizes there are still some gaps, so it'll go do some more research and then come back with like a 20-page report that it's got breaking down exactly what you asked for. To me, it's a qualitative difference because like all of you, I've got more things to do than I have time. I just wouldn't have gone and done all those things. I just wouldn't have had time. It's not just that it saves me a little time. It actually lets me do and learn totally novel things. I'm super excited about that.
What has surprised you the most as you've either built that product or you've built other products at OpenAI that you didn't expect? Because you came from some pretty exponential companies as well with what you experienced at Twitter and then also what you experienced at Instagram during the time that you were there. What surprised you about OpenAI?
I think one of the most interesting things has been that you need to build models differently in a world or build products differently in a world of AI and models because we're used to a world of semi-static, not static, but the technology that you work with, you understand. When I was building Instagram, we were using databases, and databases get better at some relatively modest percentage every year. You kind of know what you're working with. You are spending time thinking about the product you're solving and who you're building it for and how you can build them a great product. At OpenAI, every two or three months, there's a new model, and computers can do something that computers have never been able to do in the history of computers.
You need to completely jigger your roadmap around it because all of your assumptions now can be thrown out because there are a whole bunch of new capabilities. That is super fun. It's also, I mean, it means that we need to be very iterative, very bottoms up. It's very hard to plan this stuff top down. What you want is lots of smart people using the new models and trying to imagine what's possible with the new capabilities that come. It's a totally different way of building product, but it's super fun and feel very lucky to be doing it.
Yeah, and you folks are doing such a good job. One of the things we found is the composition of the team has changed where AI research now is the fourth leg of the stool. It's not just product management, design, and engineering. It's also AI research.
Yeah.
In fact, AI Canvas, when we actually interviewed the team, the AI research person had a lot of opinions on the way that this should actually evolve. Is the distance between the model and the product actually just compressing? Is that what you're talking about? It's going to be when the iteration happens, it's happening on the model, not on the product in some cases.
Yeah, and one of the things that we believe is as you build products, you want to expose the model in pretty reasonable. You want to keep the model sort of front and center in the product because the pace of change and the increase in capabilities of the model is so fast that if you try and build a bunch of scaffolding around the model like, oh, well, it's only 90% good at this particular case, so we need to build these error correction mechanisms and protect the, then in two or three months, the model is going to be way better. You're not going to have those problems, but actually your scaffolding is still going to be there, and it's going to be getting in the way.
What we actually try and do is expose the model in the product so that as the model gets better, the products just improve automatically, which is also like a little bit of a different way of thinking. The other thing that you just said was you were talking about product engineering, design, and research almost being a fourth. The other kind of cool thing that's happening is with a lot of these tools, product and engineering and design can kind of merge in ways that they couldn't before. I used to be an engineer. I write code on the weekends sometimes, but I'm no longer a serious engineer. Nobody would pay me to write code. We launched this product called Codex, which is a software engineering agent. You guys are.
We're the first design partner. Yeah.
Cisco is the first design partner. We're super excited. The idea with if you're coding today, you've got a lot of tools like GitHub Copilot, Cursor, Windsurf that are using AI to augment an individual engineer. Think of it as like a super intelligent autocomplete as you write code. With Codex, the idea is different. You actually give it a task, and it goes off and does it independently for you. You can say, hey, I want to implement this feature. Can you go do this? Or, hey, can you audit all of the dependencies in my software? If any of them are out of support, I want you to update those dependencies and then update all of the usage of those dependencies to the latest version. You give it arbitrarily complex tasks, and it can go do the work for you.
The other night, I was doing my PM things, writing emails and docs and stuff. I was like, you know, I should fix a couple of bugs.
Why not?
Why not? Yeah. I went and found a couple of bugs that looked reasonably simple. They're actually in TypeScript, which was not a language I've ever written code in. I gave it to Codex. It chugged away for five or ten minutes, came back with a pull request, and I sent that to an actual engineer. Still human in the loop, still important. Now I've submitted some code. You've got we're starting to enable PMs to write software, at the very least to prototype. You've got these image generation tools that can allow anybody to become a designer, again, at least to the prototype level. I think it means that we're all going to be much more productive. We're democratizing the ability for people to write software. Software is an incredibly generative thing.
If you can write software, you can solve all kinds of problems. I just think it's a super exciting world. By the way, we're very excited for our partnership.
No, we are actually, so one of the questions I had was, what percentage of code do you think will eventually be written by AI? And then what will the engineers do if the code is being written by AI? Talk about that a little bit.
Yeah. So here's my analogy for this. Have people seen Hidden Figures? The movie, it's about the space race and these women that played an incredible role. A bunch of that movie is you have these scenes where they're like doing longhand rocketry calculations on big pieces of paper. You look at that today and you're like, oh my God, we would never do that today. Computers are just better at doing rocketry calculations. I think five years from now, we're going to look at humans writing code, and we're going to say similar things. We just probably wouldn't do that today. Computers are better than humans at writing code. By the way, rocket scientists and rocket science is still a hugely important discipline. We still have lots of rocket scientists, and they're really important. We're still going to have engineers.
Engineering skills are still going to be super important. It's just going to be a very different world. I think a much faster, more efficient, we're going to be able to create more because of it. I'm personally very optimistic.
I actually think the brain cycles that these big brains who are engineers have can be applied to so many more strategic things because they don't have to take 30% of their cognitive load and worry about syntax.
Yeah.
Because that's something that's taken care of.
Totally. There are a lot of times when you're writing code, when you've got the architecture in place, you know how to do what you're going to try and do. In between you and that is just a whole lot of typing. If AI can get ahead of that and do it faster, amazing.
Hey, why did you decide to join the Cisco board?
I mean, I've been around technology all my life. My dad was an engineer at Microsoft all through the 1980s and 1990s. I grew up going to work with him. Cisco is just this, like, as someone who's been in technology all my life growing up, Cisco is this incredible iconic company. There's that. You look at all the stuff that you guys are showing today, talking about networking and security down into silicon, all of the cool things you're doing with generative UI, which, by the way, I think generative UI is a big part of the future of UI. We're going to have a lot more UI that's built on the fly. Kudos to you guys for being so ahead of the curve on that.
It is both this iconic, incredible company and a company that has so much of a role to play in the future. I could not be happier to be a small part of it and excited to help however I can.
We are thrilled that you're here. The last question for you is, if you were to fast forward to the latter half of 2025 and maybe 2026, what does because it's actually impossible to start predicting what's happening in AI these days because it's moving so fast, but you have a better seat than anyone else. What do you think happens in 2025 and 2026?
I think the trend that we're sort of passing through right now is people think of ChatGPT. They think of this thing where you go and you type a question, you get an answer. By the way, that's amazing, right? You can ask anything. You can use the full subtlety and nuance of the human language. You get back detailed, thoughtful responses about any topic in the world. It's amazing. We're going to see AI take us from just answering questions to actually doing things for you in the real world. It's not just going to be a thing where you go and get a particular answer. You're going to have it connected to your data, obviously with your control, but things like email and calendar and other services that you use.
It's going to be reasoning for you at all times in the background, like a super assistant would. It's going to be actually completing tasks for you in the real world. How much of your day today is something that you know you need to get it done? It's important that you complete the task, but the act of completing the task is not particularly valuable to you. It's kind of busy work, right? I want all of that automated.
With an AI that can reason, that can see, that can speak, can listen, can understand live video, can sort of interact in all the ways that we as humans interact, and can reason, so can think intelligently, inform hypotheses, and act on those hypotheses, and is connected to the services that you use every day, the tasks that you do, I think we're going to actually be able to go from reactive to proactive and automate a lot of the work people do in their daily lives and in work. I think that's a super exciting world because what it means is we all are going to be able to get more done, spend more time on things that are super high leverage for us, maybe also spend a little more time with our families in the process. That's a future I'm really excited about.
Kevin Weil, ladies and gentlemen, thank you for being here.
Hey.
Thank y ou.
All right, so we're coming to a close. All I could say to you is this is a great time to be alive in tech. Thank you for the partnership. Go to the deep dives. You'll learn a lot more about what we talked about. Enjoy Cisco Live. We are honored to be your partners. Thank you all. Take care.
Welcome back to the Cisco TV studio. We are coming to you in real time, live from the San Diego Convention Center here at Cisco Live 2025. I'm Steve Moultrie. We have just wrapped a fantastic and really empowering Tuesday morning keynote with Chuck Robbins, with Carrie Palin, with Jeetu Patel, with DJ Sampath, along with a fantastic lineup of special guests and partners and customers. I just love hearing all of these stories. I want to thank you all for tuning in with us around the world, wherever you happen to be. Thank you for reaching out to us on social media using #CiscoLive and @CiscoLive. Keep those posts coming. Next, we are headed into the first in our series of keynote deep dives. We are going to be talking AI and the secure network. Specifically, how do we power the future-proofed workplace?
Before we get to that, though, I've got some phenomenal co-hosts and colleagues. Robb Boyd and Michelle Morera are down in the world of solutions on the show floor. Hello, my friends. Hopefully, you can hear me well. Yeah?
Oh, you are coming through beautifully. For you two?
Yes, we can.
Yeah.
Excellent.
Absolutely, Steve.
What I want to do is I would love to maybe bring a couple of questions or comments to you, what we heard from the keynote space from Chuck, from Jeetu, that I thought were particularly exciting, empowering, inspiring. I want to share a couple that I thought were great first, and then I want to come to you. Chuck started out by saying the race is on. This transformation to agentic AI, it's made the network so much more important maybe than it's ever been before. We've been talking about the network forever, for decades, literally. All of a sudden, the network now means more than it ever has. He brought up some great statistics. He said 97% of us in the market today feel the push to incorporate agentic AI.
85% of us feel it is incredibly important to get there in the next 18 months. Almost none of us actually feel comfortable doing it, actually going out and getting it done, right? The onus is on us. He said your role is more critical than ever. You bring the imperative of AI to your leadership, to your board. That is kind of the good news and the bad news is that we are at the heart of driving this and making it happen. Rob, how do we think about that? How do we also do what Chuck said in terms of fusing security into the underlying network to be able to apply AI appropriately, intelligently, ethically to get to it in the way that we need to?
It's funny. I feel like it's almost like the chicken and egg problem at scale because we are having to learn about something. Speaking from kind of the enterprise or big organization standpoint, we're having to learn and adjust to something that's moving way faster than anything we've ever experienced before in our lifetimes. At the same time, we're expected to be proactive with this and kind of read tea leaves that we've never had to read before. I actually think that the push that's happening for executives from what to do with AI in general, be it agentic or generative or even previous forms, we're suffering from a little bit of just high expectations, but we still kind of need to come down to earth and say, what are you really going to achieve?
We see what we typically see in this type of thing is we see, and I'm not talking about Cisco, I'm talking about the industry in general, we see stuff that gets executed because we need to show traction in something. Meanwhile, we're going to pedal as hard as we can underneath the surface to try and make it work. I think, yes, agentic AI is the move, certainly for Cisco. I think it's important for Cisco to continue to establish the importance of, hey, this is all data-based, and our ability to network this information together and connect it is where the value is going to come from because it does us absolutely no good if the agents can't get access to the data, if the data can't be processed at the point in which it's going to provide the most value.
I think figuring out how we do that and do that consistently at scale is going to continue to be something. Yes, we all have a role to play in that. Yeah, you agree, Michelle?
I absolutely agree. Steve, I'm going to tell you my thoughts. From a high level, I loved what Chuck started out by saying that we will win together, Cisco and your company, your operations. What really resonated with me is that Cisco combines the power of networking and security together. Now, where I get excited is Splunk. I am a product marketer for Splunk Core Platform. The power of the platform pulling in data from anywhere drives that home. That is what I was excited to hear. I loved hearing it throughout each of the speakers that were in the keynote today. It just set off this day one for success.
Yeah. We're looking forward. I think in a bit further in the show, we're going to have an interview that Lauren is doing from the SOC. It's the first time we physically had a security operations center here. The analysis that they're doing, I don't know exactly how much AI is playing a part in that, but being a former security person, boy, they're dealing with an overwhelming amount of data coming in. At the very least, the ability to process that and make sense of it in real time is crucial. There's a lot of creative things that they're doing here. Steve, I don't know. Any other thoughts? Did that kind of answer your question?
Yeah, so many. Literally, Robb and Michelle, I've got pages and pages here. Michelle, based on what you were talking about a moment ago, this was an important PowerPoint for G2 that he brought up because he said, what is unique about Cisco? What do we do that nobody else out there in the market does? He targeted three things, one of which you just hit on. He started literally with the platform advantage, right? This incredible depth and breadth of capability, tech that it all seamlessly cooperates. It just works together so that it doesn't just add value. It compounds value based on the investments that so many of our partners and our customers have already made. He also talked about rebuilding into our own customized, fully programmable Silicon ASICs so you don't get locked into one specific partner or one specific methodology. You can constantly realign, readjust.
He talked about AI as a foundational principle of how we operate and build products so that it's not an afterthought when we talk about AI and how do we shoehorn it on in there. Rob, I wanted to mention one more thing that I thought was so powerful that G2 said. He said there are only two kinds of companies in the world. There are those who are dexterous with AI, and there are those who are going to struggle with relevance. I think this is no longer something that we can think about. Do we do it? Don't we do it? How do we do it? We got to go and get it done. That's why people are here at the show this week.
Yeah. I think that's basically what he's saying is some who get it, some who don't, right? We love things that are binary around here. That's how a lot of us were raised. Yeah, no, I think, and that, you know what, that holds for us as individuals as well because I don't think you have to be an AI expert. This is one of those things where everyone I meet and I talk to family, to the extent that it works, I'm like, this is something you do have to pay attention to and kind of figure out and at least be aware of the changes and kind of try to stay on top as best you can as it makes sense for your situation because it feels like it's moving fast.
I think a lot of us could potentially be left looking behind and going, oh, what just happened?
That's what I think leads into the final message of digital resiliency, using AI to get our companies, our networks, our cloud, everything resilient.
Way to bring it back home. I love that, Steve.
Absolutely, Michelle. You know what? Let's combine the resilience with what we also heard from Chuck around security, right? We need to fuse security into the underlying network. Chuck was very, very solid on that so that it's applied appropriately. It's applied intelligently. It is applied ethically in order to build proper agentic AI. I want to leave one more thing. I've got to get to the actual percentage that I thought was so cool. Something G2 said, here we go. I don't know if I agreed with the first part of it. The second part's so good. He said today 100% of the workforce is humans, but AI is going to augment. It's going to compound the capacity of throughput. Right now, 8 billion people on the planet.
The fact is agents are going to be able to perform jobs on behalf of humans, not in place of, but on behalf of humans to make it feel like there are 80 billion of us out there, really tenfold, making that specific impact. Do we have time to go back out to the show floor? I should ask our producers. Are we good with that? You know what? We are going to head over to Z. I am actually really excited for this because for some reason, I do not know how we did it. We got Matt DiNapoli here, head of Cisco DevNet Strategy. DevNet is right next door in the sales pavilion. But Z, let me send it on over to you and Matt.
Thank you, Steve. We are going to add to the conversation layer upon layer. I have the phenomenal privilege of being here in the studio with a phenomenal thought leader, Matt. Thank you so much for being here. Let's jump into the conversation. Matt is the head of Cisco's DevNet Strategy. Matt, I want to just ask you, DevNet has been endorsing the benefits of infrastructure optimization and programmability over a decade. I want to just ask you and pick your brain a little bit. How is your team embracing this new opportunities that we've been talking about? We heard about in the keynote. The new opportunities that are available, they're coming along with AI. What's top of mind for you?
We actually have been talking about AI and programmability in conjunction with programmability and automation for over a year now. I feel like we were kind of at the forefront as AI was coming up because our main audience are network automation engineers, but they're developers too. In the developer space, we're starting to layer on very quickly coding assistants. We're starting to layer on MCP services. Those kinds of things are making it easier for our technologies to rapidly accelerate. We have developers that have been doing this now for 18 months, two years, and really leveraging those coding assistants to increase their productivity.
We in DevNet saw the opportunity on top of the ability for optimizing IT operations with leveraging AIs and automation or APIs and automation by adding that layer of AI on top of that to talk about that productivity increase that was talked about in the keynote. Really great opportunities there for the people building out the integrations to accelerate their productivity.
I love that. You said two key things, ease and acceleration. That is one of the things that we've been hearing all this morning. Yesterday, you brought up the evolution of AI tools. During your session, what was really standing out for you about how this is what's changing in this space?
How quickly it's moving. I can't tell you. I feel like I'm clamoring to keep up with how broad the landscape's getting, but how quickly people are coalescing on certain concepts. Over the last three or four months, we've really heard about agency, A2A, MCP, and those things are taking portions of AI development that have been a little bit challenging for the more general technical public to pull in. It's making it more accessible and easier for people to tie in tools, tie in data sets, and actually build out these solutions that are really helpful for the IT operations people that are our main audience. Just the speed of it has been mind-boggling.
You know what? That ties in particularly to something that Shannon talked about a lot in her session about this thing of AI agents and AI application infrastructure in her session with so much focus now on agent-based delivery. How do you think our audience can cut through all of the noise? Because there's so much noise. There's noise out here, but there's also noise within their space. How can they cut out some of the noise and find practical, tangible ways to apply this in their environments?
You know, the challenge with the landscape broadening and moving as quickly as it is, like I mentioned before, is that it's hard to tell who the winners are going to be. I think that we have seen opportunities present themselves in the last really few months where the names that we hear over and over again are the ones that we can probably trust, at least in the short term. Hopefully, in the long term, they become really good partners of ours. Obviously, working with groups like NVIDIA, I mean, that's a no-brainer, of course.
We're going to be working with LangChain to build out better-together type solutions in this space around deploying agents and being able to not just have to build those out from a coding perspective, but build them into platforms that provide those low-code, no-code interactions that allow us to chain together agents and have them interact with each other by providing larger solutions. I think that's the advice that I would give. I mean, obviously, we'll see how it plays out, but I think it's really interesting to see as the landscape starts to come back together who those prime players are going to be.
You know, something that you said that really resonated that I think we need to coin and hashtag is better together. You know who else is better together? Me and my co-host, Steve. I'm going to throw it back over to him. What's going on over where you are, Steve?
Thanks to Matt as well. Just really great insights. I adore the work that DevNet does. Again, they're right next door to us here in the sales pavilion. The DevNet Strategy, being able to really boost up those thought leaders to get us where we need to go. As Matt said, it's about the speed. It's about keeping up with the pace of change and the pace of new development. It's the hardest thing in the world to do. Bravo to Matt and the DevNet team. One of the things that is so interesting and fun here at the show, when you're down in the world of solutions and you're checking out everything happening across the Cisco showcase, the one Cisco depth and breadth of our portfolio, you can actually stop in to the Network Operations Center. This year, it is sponsored by Legrand and by NetApp.
Attendees here at the show can check out the equipment that is powering this event out on the show floor in the showcase. They can learn about the San Diego deployment from engineers and solutions experts. It's very cool. Right now, Robb is down in the NOC with the fabulous Andy Phillips. Hey there, Rob.
Hey. Got to make sure I got the microphone facing the right way. Thank you, Steve. Hey, I've got Andy here, old friend, not that you're old, but we've been friends for a while here in the NOC. Andy, one of the things we always talk about when we come out here is I just always think it's fun to remind people that what we have here is not a demonstration of capabilities that we're coming out with or anything. You guys push to a certain extent, but you're all about the same thing as our customers are about, which is we need high performance reliably for a scale of people. Oh my gosh. Tell me what's new at the NOC this year. Now, how have things gone? Welcome.
Things have gone surprisingly well. Yeah, it's a new venue, which is always a challenge. We love to obviously bring Cisco technology to the customer. As Robb was just saying, everything you see behind us is real. It's live. It's what we use to manage the network. It's what provides services throughout the entire venue for every different aspect of the event, from registration to testing center to the presentations we're doing right now, which we streamed over the network. Everything is live. It's real. We love to empower people to use it. I think it's one of the best demonstrations, if not best real-life use cases for our products.
I got to think you guys get the big questions. There's always a crowd around here. As soon as the keynote let out, boy, you could tell. You didn't even have to know. You go, uh-oh, something changed. Everybody came over here. Of course, the dashboards always attract people. Some of them feel a little different. Tell me what is different about this year's dashboards.
They are. The difference this year is we've actually moved from an open-source solution to Splunk. There is a lot of kind of intricate work that goes into collecting data from the network. How you visualize it is ingesting it into Splunk, which is a platform that allows us to just take massive amounts of data, terabytes of data each day, or terabytes, sorry, of information each day and actually display it to the customer. We have heuristics. Year on year, we can use those pieces of data to model the network and actually see if we need any upgrades or anything around the venue. It is very useful.
I get this feeling. I mean, of course, I know this to be the case. I consider you guys very transparent. You have sessions that you encourage interactivity. I believe at these sessions, it's some people from the team, both Cisco as well as partners that are integrating the technologies. What's the collaboration like here in terms of who you work with to pull this off? NetApp, obviously.
NetApp is one of our partners. They've been for years. They actually help enable our FlexPod, which you can see here. This year, we've actually got AI in our FlexPod. That was a key change that we've run. As for the teams, we use different parts of Cisco CX. Our company that does all our professional services within Cisco, we have members of sales, members of tech, technical assistants.
I didn't know you had sales in there too.
Yeah, we have sales too. Actually, one of our best is from sales. So he's amazing.
That's always good to hear. Yeah. I know, OK, you mentioned AI being in here. Then we talked about, I was never remembering exactly how we referred to it. It was AI Pod? Is that the right way to put it?
AIPod, yeah.
What exactly are we doing with that here?
Actually, AI Pod means UCS system that's actually enabled with AI. FlexPod AI is our partnership with FlexPod or NetApp, sorry. In this case, over here, what we've got is the FlexPod is the accumulation of storage, compute, and network. It's an out-of-the-box kind of data center solution. What we've changed this year is you can see actually the second blade on the UCS X is what we call a PCIe node. That blade actually allows us to install GPUs, in this case, NVIDIA GPUs. We're using that for all kinds of AI applications in the NOC, some for us to test and see what we might be able to do.
That's a big enough channel on the PCIe?
Yeah, it's huge. Actually, it's shared. There are two blades adjacent, and the blade adjacent to that node is able to, across the X Fabric module, which is a PCIe bus, share and use those GPUs. You could have a virtual machine that can use the GPU. We have four across the whole data center, so we've actually got four virtual machines running with full capacity from that GPU. Yeah, we're running an app, actually, which is in test right now from our CX organization. It ingests all of Syslog and actually is able to inference what problems we might have on the network. During that time, we're actually getting a report that says, hey, this is what we think from the AI's perspective is the top kind of things you should be concerned about. We investigate those.
We actually see things that maybe we would have missed because Syslog is so intensive. There's so much going on. A human can't really look at all that data.
That is just what I was thinking about too because I was like, oh, what I heard you say is less or no log reading, which is great because there is so much that comes through there. I feel like that was probably the first thing we all thought. We saw generative AI and the ability to parse data in a way that makes sense to us. Obviously, you guys put a lot of guardrails around that. You have had some access to the AI Pod to ask questions and stuff over here, I have noticed as well. Has that been well received? What do you think?
Yeah, we're looking at statistics of how many times it's been used. It's been used quite a lot. Obviously, our biggest concern was, hey, we don't want people to ask questions that might send something we don't want to say. We've actually locked it down. It's really just looking at Cisco's external documentation that's public facing, but specifically around what we have in our NOC.
We just got about one minute left. I wanted to ask you because I know being in a new facility and such, you guys were working with them. They purchased wireless gear.
They did.
You guys have kind of helped with the installation and making sure it meets all of our needs. These are very high needs, which obviously benefits every customer they have in here after us. I'm curious, this is the first time I've seen us really do. We talked about it, but the location awareness at the BLE that's turned on now with that. Can you talk a little bit about how's that been received? Is it a nightmare for you guys?
We have to say thanks to the venue firstly because they actually installed an extra antenna on all of the APs. That gives us BLE. The Bluetooth location analytics is much more accurate than standard Wi-Fi. It has given us, if you use the Webex board strewn around the venue, you can actually get a point A to point B direction.
Yeah, I'm very impressed with it actually. It's always worth the bathroom.
Yeah, it's pretty good. It's my next thing.
No, I love that. I like how you guys make such incremental improvements. I'd love to go into, but we're not going to have time today, about how you guys decide what the cutoff is and what you deploy and how you plan for that. It's always save something for else, save something for later. Speaking of that, however, we have purposely avoided talking about security because Lauren is going to be talking to security in just a moment. With that, Steve, that's just a tease. Steve, we'll go back to you.
Thank you so much, Rob. I appreciate it. You're right. Always leave them wanting more. Andy Phillips, one of our favorite people to talk to. Thank you so much, Andy. We'll be back more with him later. We are going to take a quick shift right now. Z is very close by, hanging out with one of our special guests here. Daniel McGinniss is with us, VP Product Management with Compute Cisco. Z, let's send it over to you.
Yes, Steve. I'm in the studio with two amazing thought leaders, Danny and Sean. Go Trojans.
Absolutely.
Dakota State University.
Very good.
Hello and welcome.
Hi, Z.
Excited to hear the conversation between the two. Danny.
Yeah, sounds good. Sean, thanks. Thanks for being here with us. Appreciate you coming to Spotify. You guys are phenomenal partners of ours. I think we're doing some great stuff together. I guess you know you're here. Customers want to hear about what you're doing. Talk to us a little bit. There's a ton happening in higher ed right now, I think, in just the shifts we've seen. Talk to us a little bit about what you're seeing from the university standpoint, change in the industry over the last period of time.
Sure, yeah. You know, from my perspective, the importance and the impact that research has had has historically been overlooked. I think there's multiple reasons for that. One being that with former technology capabilities, the speed to get data results from your research, it took a long time. That cycle was long. It took a long time to set up the research and to do the testing and all the analysis and all that. That data that you got probably had a specific function or use case at that time, but eventually became potentially a liability to the organization. You archive it and set it aside somewhere. It's just there. It just builds up. There wasn't the ability to index or use that data readily.
Certainly with the innovation and improved technologies that we see, specifically AI, we really have reduced the amount of time it takes to do that research, to perform that, to set it up. We have results in mere days and hours when it used to take months and years. That is really a value add. That data is readily accessible. With AI, which has shortened the time to do that research, you also have all your data is usable at all times. That just, again, adds value to that data and being able to utilize it not only for what your purpose is now, but for language models later. That is really a change that I've seen.
Yeah. I mean, I think, look, you hear everybody's talking about the importance of data. Data is what it's all about. I think your industry is very similar to what I see across a lot of the industries, especially when it's about consuming a mass amount of data. We're even dealing with a very similar at Cisco and how we're transforming. All right, now take it down into the technology stack. You've been running IT here for quite some time. I mean, obviously, there's a whole new level of complexity. There's new performance demands. There's new architectures you're dealing with. How are you and the team embracing that and managing it?
Yeah. We certainly, you know, there's certain items that we look at for everything. Cybersecurity is very important to us. Scalability, observability of your systems, of your data is very important. For us, centralization of our services, simplifying those workloads and the cybersecurity that we perform, not only for the systems, but the data across that, having full stack data observability with Intersight and with all the platforms from Cisco has immensely helped us to reduce the amount of time it takes to manage the environment, but also improve our cybersecurity posture. It's just a great benefit to us, simplification.
That's great. Yeah, that's great. Sean, you guys are doing some great work over there. I really appreciate you being here and spending the time with us, especially the partnership. Thank you very much.
Thanks for having me.
All right.
Thank you, Danny.
Thank you, Z.
Thank you, Sean. What an impactful user story. Steve, what are your thoughts on that?
You know what? Always brilliant, brilliant content coming out of the studio, out of the keynote space. We're going to make a very, very quick turnaround right now, though, Z. A few minutes ago, we had Robb and Andy in the NOC. Right next door to the NOC, we have got, big surprise. The SOC is there. Our attendees here at Cisco Live, they can get an exclusive behind-the-scenes look at how active threats are being detected by Cisco Security and Splunk in real-time traffic. Lauren is in the SOC right now with Jessica Oppenheimer, Director of Security Operations. Lauren, how's it going out there?
It is going amazing. Hi. I am here, like you said, with Jessica, the Director of Security Operations here at Cisco. Jessica, this is the first time we have had a SOC at Cisco Live Americas.
That's true.
What is the mission of the SOC?
That's a great question. As you said, it's the first time at the Americas here. Obviously, you've talked about the Network Operations Center. Like our customers who evolve in their enterprises, they start out with security in the NOC, right? They say, hey, we need network security, the firewalls. You're going to do that and maybe introduce domain name service security like Umbrella, secure access, and secure network analytics for the network detection and response. As it grows, they realize they need to mature. In the past, we had something like this, which is the SOC dashboards, which we still have here. If you came to Cisco Live Americas in the past, you would see the SOC dashboards. One of the folks that was really smart in it would come in to answer questions and talk to people about it.
It was like an engagement. There was not actually anyone behind the keyboards looking at the data and searching for threats. Now we have that. Our core missions are to protect, to educate, and also to innovate. That is where we really focus our times on. Let's talk about that.
Wow. I just want to know what prepared your team to be able to take on this.
Yeah, as you can imagine, getting ready for a Security Operations Center at a major event like that, there's over 20,000 people. We just came from the RSA conference where we've had the SOC there. I've been leading that for Cisco since 2017. Some of these team members were protecting the Super Bowl just recently, Black Hat. We were at Black Hat Asia. We had a lot of innovation. Team members like myself have been to Paris 2024 Olympics. I've been working on this for 10 years.
Wow.
It was a passion project in addition to my other duties in XDR. Now it's my full-time job as well, working with our partners like CrowdStrike, Palo Alto, other folks that we need to integrate with and have those relationships because we can't do this alone. Our customers are going to have a stack in their security that includes different technologies. We need to make sure that all those work together. That's really a focus of what we do in SOC work.
That's the thing. I feel like so often we think that we should be siloed or single vendor approach or anything. At the end of the day, it's one platform that secures. Speaking of those different technologies you named, what technologies are we using to protect Cisco Live and the network here? Even our camera team and all of us, what are we doing?
Yeah, you know, Cisco TV is our primary customer today. Make sure that's great. Why don't we go into the SOC and we'll talk about the technologies because sometimes seeing right there is best. Let's go inside and we'll have that discussion.
Awesome.
We've had over 550 people come to the Network Operations Center yesterday and look through the glass and talk about it. We've also brought people inside for tours, and we've had over 100 people do that. One of the things we talk about is the topology and talk about the technology. At the core, we're getting that span from the NOC, the Network Operations Center. As we've discussed, you can have a NOC and a SOC together. At Black Hat, they're co-located. They're in one space. As you grow an organization like this, they will be really next to each other. We have technology, what's called a SOC in the box that Andy Audie here, excuse me, he put together, one of our engineers.
That allows us to deploy a mobile SOC anywhere in the world, plug it into the Network Operations Center like we did when we arrived on Friday, and then connect that to the Cisco Security Cloud and the Splunk Cloud as well with enterprise security. Before all that, when the packets flow in, we go with InDATES, which is our core packet capture technology partner. They've been a partner of ours for many years. Every packet that flows through, we're able to record that. With those integrations that we have from our XDR team of analysts, when we're looking at an incident, they can right-click and go right to the packets when they need to. Also from Splunk, which we see are the different dashboards and our firewall.
That is incredible. Of course, like me coming Cisco and Splunk, I love seeing it all and seeing us be better together because we truly are. I know we were laughing and kind of not laughing because it's not a joke, but it's been a lot of things we've been seeing in Cisco Live this year where people are not using best practices, right? Not doing things as secure as they could with clear passwords and all of that stuff. Can you talk more about what we're doing to kind of prevent that and see that?
Yeah, you know, that's our first mission is to protect, one, protect the network. Some of the ways that we do this is we want to ensure that folks aren't using this network to attack outside, to attack the rest of the internet. That's a liability to Cisco. It also doesn't look good for what we're doing. That's one of our core protections. Also to protect the network from attackers within, someone that was trying to do denial of service or disrupting the cameras or Cisco TV while it was going on, or they might come infected and start trying to propagate, right? We have protections in place. If someone accidentally clicks on something, that will stop and block it. Also we're actively looking for infections that could have an issue.
Some of the other things we do is that when we have this full packet capture that we talked about, all of those logs are flowing into Splunk. We have automation to detect when someone has their password in the clear, such as their email. A lot of people are still using POP3. This is the worst POP3 we've seen in 10 years. I've got my partner in crime here, Mr. Fink. We've been working at Black Hat for 10 years together. We see POP3, email in the clear. We were just at RSA conference. We could track people down and help them individually. We started doing that here, but the volume is so much that it was overloading our XDR analysts with the amount of passwords in the clear. Our Splunk experts over here, Austin, as well as Tony, introduced automation.
Love it. Before we go here, because we have a couple of seconds now, how can others learn more about the SOC?
Yeah, so we have tours. There's a sign outside for those that are here. Also, if you want to learn more and you're watching, you can Google my name, Jessica Oppenheimer, Cisco blogs. There's six pages of blogs there. We got to go because the LA 2020 Olympics is coming in for a tour.
We're going to send it back to the studio because we got to go as well. Thank you so much for this tour, Jessica. Boom.
Appreciate it. Great job, Lauren. My thanks to Jessica and the entire team down there. So great to get those Splunk insights. This again is why it is so important and so interesting and so dynamic to be here in person. It's great to attend the event streaming live. We'd love to have you with us, reaching out to us and bringing you all the content that we can. When you're here in the space, that's when you get to go and tour the SOC and things like that. Right now, we are headed into the first in our series of keynote deep dives. We are about to take a look at AI-ready secure networks. How do we power the future of the workplace where AI is transforming campus and branch networking? It is all about dramatically increased traffic.
What do we do to push the existing architectures beyond their limits? The reality is the current network fabric can't keep up with demand, right? Security, operational needs of an AI-driven enterprise. You heard this morning in the keynote that Cisco has just introduced a new AI-ready secure network architecture. Very excited, specifically for the scale, the speed, the complexity of the AI workloads. Includes low-latency wireless and smart switching and secure routing, high performance. We're going to head out into the space right now. We're going to hear from our VP of Product Management. Here we go.
Please welcome Senior Vice President and General Manager, Network, Platform, and Wireless, Lawrence Huang.
All right. Good morning. Welcome, everyone. Welcome to this morning's keynote deep dive on how we're going to help you future-proof your workplace. My name is Lawrence. I'm the General Manager of the Cisco Wireless and Network Platform business. I'm excited to go deeper than what you just heard in this morning's keynote. As you heard from Jeetu, we're coming together across as one Cisco really to deliver incredible outcomes that our customers tell us they need help with, whether it's modernizing your data center, not just for the workloads of today, but for those AI-powered workloads today and well into the future, to digital resiliency across the complex ecosystem of your digital infrastructure, across that owned as well as unowned infrastructure.
Today, we're going to focus specifically on how Cisco can help you modernize your infrastructure to future-proof your workplace not only for this year, but well into the next decade here. We have a unique opportunity here at Cisco. You already know that we have a wide and deep portfolio across networking, security, collaboration, and observability. Sometimes the sum, the total, is not greater than individual pieces. The reason for that is we haven't always taken a platform-centric approach to the way that we build products, the way we deliver outcome for you that you can deliver to your end users here. At Cisco, as part of the way that we're doing things differently together, we are taking an AI-centered, platform-centered approach to how we go help solve the problems that matter for all of you here.
Of course, through all these investments in viewing everything that we do, not with AI as an afterthought, but in viewing it into every single capability that we think of bringing to the market for each and every single one of you here. Now, why are we doing these things? There should be a good reason for it, right? It's not just because it feels great, but it's because we are living in times of change here. Just a show of hands, how many of you were around during the internet transition? Mobile transition. All right, great. I think, like many of you, I'm a little bit older than I would like to be. But I have lived through each and every single one of these transitions.
Time and time, what comes across as surprising, but not so surprising, is it all feels like it comes out of nowhere. Then it just accelerates so rapidly, right? The pace of change only increases while we are sitting in this moment in the next transition. While they say that history doesn't repeat itself, that it rhymes, the rhythm is telling us that with each and every single one of these transitions, the things that we learn is that you need to rethink how you build, design, and deliver the infrastructure and outcomes for your users. This moment is no different. At Cisco, we believe that this moment requires us to rethink how we build infrastructure for the age of AI. Really, this shouldn't come as a surprise to any of you.
The trends that many of you have been telling us about, the challenges that you have where you are expected to deliver more for your organizations with less, how the scarcity of talent isn't getting better. If you were in this morning's keynote, you heard how the proliferation of AI means that there's going to be more AI agents out there doing machine-scale work than there are humans to go tackle that. Ultimately, we have to figure out how do we actually build our infrastructure in a way to help respond to these changes that we're seeing in the broader marketplace here. Rather than have me speak for the whole session, I do want to bring up some special guests.
Really, what we're going to be talking about is how Cisco is investing in three major areas to help do just that, how we're helping build new network devices, not just simply a refresh, but really network devices that we think are going to be future-proof for the next decade and beyond, how we're infusing security into our infrastructure. Of course, our perennial favorite, the evergreen thing that many of you ask for, which is how do you actually simplify your network operations? Because we know this continues to be a challenge here. Let's get into it. I'm going to welcome up our esteemed panel here, Austin, Lee, Lizzie, and Nick. Please join me on the stage. All right.
We have an esteemed panel across many of the exciting announcements from our network platform, how we're bringing the platforms together, to our secure routers, to our campus switching technology, and of course, our investments in AI. Maybe we'll start a little bit at a higher level here. AI has taken center stage. It is not surprising that this is the topic on everyone's mind these days. I'm curious to hear from this panel, how do you think our customers should think about what this means, not only for their organizations, but for their infrastructure? Maybe, Lee, I'll start with you.
Yeah, so certainly, we're known as a hardware company, and it starts with building great hardware, right? That's our big focus on making sure we've got amazing devices. More and more, you'll see the focus consistently across the company around how do we ensure that those hardware devices are not standalone, that we have a broader platform. I know Austin will talk a little bit about that as we go along here as well. G2 loves to use the iPhone analogy, and I think it's a fantastic analogy because we look at the iPhone. It's an amazing piece of hardware. It has a really powerful OS, and it has services from an iCloud perspective. They're all intrinsically linked. We don't separate those pieces in our head. We think about the sum of the parts being greater than the whole, right?
That's really where I think we're focused on is making sure that these amazing devices have amazing management plans on top of them, very powerful ways for you to address and integrate and implement them and just make sure that they're very easy to use. You'll see a lot of ways we've come together under G2 to do that in a much more consistent way than we have in the past.
That sounds spot on. Maybe, Lizzie, I know your team has been focused a lot on how do we drive more simplicity into everyday operations. What do you think about this?
Yeah, so I'm working on the software side. When my team talked to customers, we heard a couple of themes of how do we make things simple, like how do we scale across multiple data sources and bring them together, as well as how do you hand off, right? Many people tend to work on networks, right? How do you have a single source of truth where you can look when there's a problem or all look at the same page? Lastly, looking at purpose-built AI is great, but also a concern to keep the human in the loop, but make it efficient, right? Those were the themes that we heard when we were looking at research and our product development.
Yeah, I think this idea of keeping the human in the loop, building trust, this actually makes me think of this idea as we build AI and it becomes more pervasive in enterprise operations. Maybe I'll ask you this question, Austin. How should this audience think about the architecture, the security, all this without compromising on trust? Because trust is such an important part. We know that the feedback from customers is they want to have trust in the systems that we're building.
Yeah, absolutely. I think that you've heard a lot from Jeetu and even Kevin from OpenAI and that the workforce and what we think about as work today is foundationally going to change in the next few years. This change is going to happen faster than we think, where now all of us are leaning on AI in our tools every day. Our businesses are pressuring us more. Your business, our leaders are looking for more demonstrable progress towards productivity gains and how we actually show ROI on that. All of that comes with new ways of work, whether this is AI assistance or agentic workflows.
In the future, you could very easily see how we're going to augment our workforce with many, many AI agents doing many of the things that might be more repetitive or things that we can use to augment our teams. Now, with that, that comes with new challenges from a security perspective. Making sure that the AI agents, you know what's happening in your network. You can see the types of traffic, the communication. You can actually do the segmentation to actually secure your network, as well as protecting your network perimeter from new types of attacks that are AI-enabled that will affect our networks in different ways than we see today. Beyond that, I also think that it changes how we operate as IT leaders and IT operators, where we have to think about more ways we can scale our own teams, right?
This is where AI embedded directly into the network allows you to pinpoint issues, to move from GPS to self-driving on your network, right? Not only guiding you to where you find issues, but being more autonomous to actually spot and prevent issues from happening in the first place. These are all technologies that we're deeply embedding in our operational models to help you be more productive as IT teams.
Yeah, so it feels like it's like layering on levels of trust and increasingly over time, more automation as that trust is earned.
Exactly.
We've talked a lot during the keynote, the lineup of new products, the refresh products, especially built in the age of AI. Maybe I want to transition to a little bit about this. We talked a lot this morning about this idea of smart switches. It's a category of its own, right? That's what G2 said. Nick, maybe I'll come to you. Is the idea of smart switches, what exactly is it? Is the word smart and hyperbole? Tell us more about that.
Yeah, I think with smart switches, it starts first with the customer. Customers need switches that can still deliver all this high capacity, bandwidth, throughput scale, low latency, but they're also looking for the network to solve additional problems and use cases. That's where the smart comes in. There's additional compute on these switches that have nothing to do with the forwarding plane and control plane that gives the ability to deploy capabilities for better telemetry, for observability, better logging management, integrating with Splunk, and ultimately additional AI capabilities that can reside resident on the hardware and future HyperShield readiness. I think it delivers more capacity that augments not only the forwarding and the throughput, but addresses additional use cases powered by the network.
Yeah, I really like this idea that it's not just building the next-gen switch for a higher network throughput, but building in capacity to build programmability into it. Because ultimately, I think it's true that no one in this room can truly forecast what's coming over the next few years here. I also know the industrial IoT team has been busy. It wasn't just about building new products for the enterprise environment. Nick, do you want to touch upon that a little bit more?
Yeah, so the IoT environments, the industrial world is moving as rapidly as traditional campus incorporated office space with more demands for the devices, IoT, autonomous vehicles. I was talking to a customer this week where they have these deep mines and they have these boring devices that are going through there, and they're transitioning those to just automation, fully autonomous type of vehicles. With that, they need the same things that we need in traditional carpeted office space. They need high throughput, high scale, but they also need the ability to address the flexibility required in these environments where low latency is critical and some of the Wi-Fi use cases may not address everything they need. With the ultra-reliable wireless broadband, it gives the backhaul, it gives the ability to have better functionality all in one box because it combines Wi-Fi and the URWB.
There is one device that gives the operators maximum flexibility, and it really addresses their use cases as they look to future-proof their own investments in the industrial environment.
I got to hand it to you. You almost got through the curve acronym without stumbling.
I was programmed with curve. I was ready, yes.
Yeah, absolutely.
Also, we've done a lot of Wi-Fi 7 announcements here. So what does that mean for the large venue type of customer profile?
Yeah, you know with the wireless portfolio, you're asking me questions. Okay.
Back up.
We did announce Wi-Fi 7 back in November, but at this Cisco Live, we also introduced the campus gateway. Those who want cloud management at campus scale, solving use cases like fast layer through roaming, that's all now available. I think the large venue and the stadium access point, it really is the industry's first Wi-Fi 7 access point for the enterprise for those types of deployments here. Of course, building in a lot of programmability into the radios to provide custom beamforming depending on the type of radio configurations you need. These are things I'm excited about. Okay, maybe I'll go back to the panel here. Lee, secure routers. Aren't all routers secure?
It's extra secure. A few things there. There are two big trends we see universally. One is that security is everybody's job now. I think gone are the days where you were just the networking admin and the security is somebody else's problem. More and more, our customers are saying we need to have an integrated solution. Perhaps NetOps and SecOps are still two separate personas, two separate operating identities, and we need to be able to address both of those. It's obviously much better to do that in a single box. Less equipment to install, less things to go wrong. That's sort of trend number one. The other big trend is just the explosion of both the availability of bandwidth and also the need for that bandwidth. I tell people this story to usually some jealousy.
My place in the Bay Area has a 10 gigabit per second symmetrical fiber connection for $40 a month. I remember getting my first 128 kilobit per second DSL and just thinking, well, this is so much speed. Why would I ever need more than this? The speed is available, but also the need for that speed, particularly as we're getting into branch locations now that there's more AI getting pushed to the edge and lots of applications there that are very bandwidth hungry. Also, we need to be able to do this in a way that's consistent from a latency perspective, from a reliability and survivability perspective as well. Those are the two trends as we think about how do we build the right devices for that.
Now, to hit on the points within this actual product itself, we're introducing within this next generation of products. It's an evolution of the current product. You would expect the generational improvements that come with that in terms of speed and throughput. That's all done by what we call the secure networking processor in this device. That's a piece of silicon with Cisco intellectual property in there, solely focused on how do I accelerate cryptographic workloads. That has a number of benefits. One of those, though, is it allows me to run NGFW firewall services on this box without a massive penalty from a throughput perspective, being able to do the inspection there in the device itself and have the firewall rules apply within this box. It also allows me to be ready for a post-quantum cryptography world.
Imagine a quantum chip comes out and its ability to parallel process essentially obliterates any of the techniques we're using today to secure either the devices themselves, data at rest, or data in transit. Being able to actually run higher order encryption on these devices does need a special type of processor, and that's going to be available in this product as well. Finally, we need to manage this, right? The other big innovation here is making these devices able to be managed via the cloud. We're following the beauty of the platform in terms of how we build from an operating perspective is that Nick and the switching team have done a lot of great work, as have the wireless team, on getting these devices manageable by the cloud.
We're able to inherit that work because we use the same underlying operating system. The third of that, Jeetu talked a lot about the security cloud control today. What we're able to do with that is have that be the single point of policy, sorry, policy definition, and then distributed enforcement, including these devices as well. You can have a common policy that applies to cloud devices, applies to the Cisco Secure Firewall, but also applies to the Cisco 8000 Series Secure Router that we're launching here at Cisco Live. These are available now. You can order them, shipping in July and August.
Lee, I think you may have a customer video to bring this home, right?
Correct. We have lucky enough to have one of our amazing customers, Nestlé. Nestlé's mission is to use the power of food to enhance life, which really resonates with me. We are very honored that they trust us to power their business as they go pursue their mission. Why don't we throw it to the video?
Could you share a little bit about the size and scale of Nestlé in your business?
It's massive. We're talking about 275,000 employees around the world. We're talking about 188 countries. We have more than 300 factories. All of them need to be connected to support our business every day. At Nestlé, we are not adapting to change. We are engineering for it. As we enter the area of agentic AI, I believe that complexity will only expand, demanding even smarter, faster, and more resilient networks. With Cisco, we've built a network that's agile, resilient, and ready to meet the demands of the world's largest food and beverage company. Our factories are not where the networks are. Our factories are where our raw materials are, where the farmers are. We have come up with solutions in the branch office where they just have to plug it in and it works.
Anchored by Cisco's next generation secure WAN solutions, including advanced routing, a unified SD-WAN fabric, and a global cloud backbone, we can dynamically route traffic across multiple clouds and providers. With Cisco, we've eliminated MPLS. We manage everything securely from a single dashboard. This gives us centralized visibility and control across thousands of sites without needing multiple point tools. I remember when a major undersea cable was cut in Africa. Other businesses were offline for weeks. We were back online just under a day. Today, we are maintaining 99.97% availability. Our mean time to detect and resolve issues is at record lows. Cisco brings it all together and makes sure our heart is beating every single day.
That's great. I always love hearing directly from our customers. We'll definitely bring up more customer examples in a bit. Lee, you mentioned this idea of quantum. Nick, I know with the smart switch announcement, one of the things I heard from your team is this idea that we want to be ready for the post-quantum compute world. What does that mean? How should the audience think about that as a potential risk that they need to think about for their infrastructure?
Yeah, I mean, I think at a high level, the risk is that adversaries and nation-states are looking towards the notion of harvesting traffic now, storing it away. Then when quantum computers are ready, they'll decrypt and they'll have all the secrets of the known universe, all this kind of thing. Now, the reality is that these customers who are looking at this problem space, they really haven't had a standardized way of addressing this problem. NIST and standards bodies are now just getting to the point where they're ready to put pen to paper and put these things in ink. As we think about our switching portfolio, our routing portfolio, our hardware portfolio, customers look to make long-term commitments and bets with these products. The products that we're shipping are going to be PQC ready when this world evolves.
Starting at the hardware layer, all the way up to connectivity, control plane, and transport, these products are going to have this ability so customers can make this bet knowing that as it gets ready and ratified in EU and America and APAC and all these places, that their investments in hardware will be ready for them as they get these deployments out in the world.
Yeah, I mean, I think that's interesting. I mean, this is one of those topics where I don't think a lot of our customers think about this on a regular basis, but it feels like this could be one of those step function changes for the industry when it becomes front and center here.
Yeah, yeah, 100%. I think as we look at the switching and the routing and the overall network infrastructure in general, networking has been a way that we help customers solve security problems. We refer to that as secure networking. This has been something we've done for a long time. The bad guys know this. They're attacking the infrastructure to get in. Yes, they'll still do the phishing attacks and all these things and the downloads, but they're really hitting the infrastructure itself. In 2020, there were roughly 18,000 CVEs published. In 2024, there are over 44,000. This volume is going up. If the network infrastructure is doing the watching, who's watching the watchers?
What we're introducing, and you heard Jeetu talk about it today, is this notion of Cisco Live Protect, which is a technology that will look at the product itself and provide real-time security as we understand that attacks are coming in targeting the infrastructure. Maybe we can cue the demo here, and I can kind of start to walk you through some of the problem space. Essentially, as you look at your infrastructure, I was talking to a customer last week about this. When they try to solve these CVEs that get announced, security team's involved, networking team. They're trying to look through all the networks they have, what devices are impacted. First, we are leveraging the context that exists in the controller, whether it's Catalyst Center or the Meraki dashboard. We're providing that context to the customer directly.
What the CVE is, what the CVSS score is, are their products susceptible to that? We're giving them the ability to then manage this workflow themselves. As they click through, we show all the context associated with the CVE, what the scale is, what the impact is, what it's addressing. We're going to give customers the ability to deploy a compensated control to protect that device from that particular issue. Now, the alternative is wait 45 days, download the patch, upgrade your network. That requires time and then downtime in the network. Sometimes you might still not understand exactly the scope of what's being impacted. In this scenario, you can pick the specific device that needs it. You can deploy it.
We give you the option of just putting it in observation mode so you can see what the impact is if someone is trying to trigger that particular CVE, or you can deploy it in full protect mode. You can have it in your network ready to go. The thing that's really critical is with this technology, we can then give this full visibility into what is actually attacking that network device. Part of the Cisco and Cisco advantage is you can use this type of tech to not only protect your devices, but then you can follow it all the way through to see it looks like there's a compromised endpoint in my environment that is actually attacking this particular piece of network infrastructure.
All this logging will be resident in Catalyst Center or Meraki, and it can also go straight to your Splunk instance. You can have that type of visibility and control. It is one of the things that we're doing to really try to upgrade the overall defense posture of our customers' networks as they rely upon this infrastructure to deliver better security for applications, users, and devices.
I really like that. I don't know about smart switches. Security switches, I definitely buy. This compensated control thing, that's really amazing. I think about the stat G2 shared, right? The 45 days on average it takes to patch something, the three days to exploit the vulnerability. I buy it. Hey, I also want to transition a bit to this idea of unified branch. I mean, Lee, don't we want a unified branch? I thought we already had it.
Absolutely. This is when we talk about the power of the platform. The idea that all the products we announce work better together and they're able to operate very seamlessly. We do that by taking that very platform approach to how we build the management layer and what sits on top. What that allows us to do is a few things. One is that that's a consistent area where we can apply intelligence. Being able to run AI, whether it be specific to the individual domain you're troubleshooting or some of the stuff that Lizzie's working on in terms of doing that across multiple domains, it's very important to have consistency there. The other piece where we're putting a lot of investment in right now is how do you as a user interact with that platform? You can use the UI.
The UI works great. ClickOps, as we sort of refer to it. What we're developing and announcing here and developing documentation around, we're using the Cisco validated design process, is the idea of how do I use infrastructure as code to interact with this platform and be able to do things at the right scale, with the right level of security, the right level of consistency, both from a deployment perspective, but also from an ongoing management perspective. Leveraging the power of Terraform as an open standards-based way to do this. We have a lot of material that you can pick up.
If you're already comfortable with DevOps and operate yourself, but also working with our partners at CX can help you do this as well if it's a practice that you want to get up and running and being able to leverage our partner community to go and essentially use infrastructure as code to go deploy and manage the network and network configs the same way you manage software.
I love it. CI/CD type model, modern software development workflows, this is really how you help scale these branches in today's era.
Absolutely.
Yeah. Okay. Why don't we transition a bit? Austin, there was a lot of focus that Jeetu talked about, hey, we're bringing together our platforms. We've been talking about this for a while. Where are we exactly? Tell me what's real.
Yeah, absolutely. Even today, we have the industry's most highly used platforms in both Meraki Dashboard and Catalyst Center. We've been on this journey to actually bring these platforms closer together to deliver a more unified management experience. We've had a number of amazing accomplishments and proof points that I'm really proud of that the team has been working on, starting with things like the user experience. We've unified on the look and feel of both platforms. We've started working on actually converging the backend system so that the AI systems that power things like RM on the wireless access points are powered from the same machine learning backend, the same AI/ML models on both platforms as well.
We're going beyond that with delivering things like a consistent marketplace, a consistent API, as well as unified hardware and licensing across the whole portfolio now with all the next-gen products. Beyond that, on the management experience side, you're going to start to see more of the workflows that bring the power of Catalyst to the cloud, as well as the simplicity of Meraki to Catalyst Center as well. A great example is the Live Protect workflow that Nick showed, but also things like bringing large campus switching features to the cloud, things like VRFs and BGP fabric we're bringing to the cloud, as well as bringing more simplified workflows to Catalyst Center. There's a number of things that we're super excited to announce this week and continue rolling out going forward.
I love it. Really a lot of enterprise-class features that many of you in this room rely on, now bringing it to the cloud management model. Okay. Hey, the AI assistance and this idea of agentic ops, we've been doing a lot of work here. How does this fit into the broader vision?
Yeah, absolutely. I'd love to actually illustrate that with a demo. If we could switch over to the demo here. We've made a lot of capability improvements onto the cloud, like I mentioned, bringing large campus capabilities, but also enhancing this with powerful assurance and AI ops capabilities. Here you can actually see our brand new organization summary page that gives you a guided perspective on where you need to be spending time in your entire deployment, what's working, what's not, and really drill down very quickly to any hotspots or where you need to be taking action. You can actually dive here directly into a client view, see if there's a client that's having issues. Cisco is the only provider that can actually look from that client all the way out through the LAN, through the WAN into the applications.
This is powered through the ThousandEyes integration. You can see that entire path visibility in a single place to really pinpoint and diagnose where that issue is happening. Not only can we do that, we can actually look historically at that client to see if there's any kind of connection issues or any failures that have happened. You can actually go back in time and pinpoint, here's a time where one of my end users called in, here's the type of issues that they were seeing. Maybe it's connecting to a Wi-Fi access point was the issue, but we're actually able to do the heavy lifting with AI for you.
We're actually doing the correlation, looking across all the things that happened on the network, including any changes that you've made on your network, and pinpointing, here's actually a change that we believe is correlated to this outage. Really pinpointing that and doing that heavy lifting for you. Now, you can diagnose this in the UI, as well as if you want a more guided flow or say that you have a help desk or a NOC team that really wants a more guided workflow, we have a bunch of new capabilities in the AI system that we're launching this week as well. Here you can actually go and you can ask, help me find the networks that are having issues, whether it's authentication, whether it's other issues that you want to diagnose on your network.
The assistant will go out, fetch all the data, provide a very simplified answer to you that's summarized for your role. From here, you can jump off into troubleshooting flows directly from the assistant. You can click troubleshoot. The assistant is actually going across all the events, all the logs again, providing you not only what's happening, making sure that we continue to give you the control and visibility of what's happening in your network, summarizing any kind of actions for you, as well as giving you some recent changes. Here's really important context on what happened on the network that otherwise would have taken maybe multiple round trips or multiple log sessions to actually understand what's happening.
You can really quickly follow up with more detailed questions, so understanding who made the configuration changes, what the other things that happened around that network around that time. The assistant really gives you a comprehensive answer to that as well. One of the great amazing parts of the assistant is actually it blends not only all the network telemetry that we have, but also things like our best practices, our CVDs, our documentation. For customers that want a more guided workflow, you can actually pull in that knowledge as well. That provides a very summarized answer. Here I can ask, how do I actually go avert these policy changes? It actually pulls in documentation, gives you a very clear, here are the steps that you would go to summarize and take action on that to resolve that issue.
Hopefully that shows you an interesting example of how we can actually take a diagnostic flow that might have taken maybe hours to trace down from a wireless issue to actually a policy change and turn that to just an order of minutes.
That's great.
Yep. I was going to say this is just we're just getting started here, and this is kind of the foundation for how we're going to advance this with things like the deep network model, as well as more agentic workflows.
Yeah, I really like this idea that you showed even in the troubleshooting scenario, this idea of continuing to build trust in why the model's telling you why they discovered an issue. I also noticed it happened to look like dark mode, by the way.
Yes, dark mode is coming very soon.
All right, fantastic. Hey, Lizzie, this morning, G2 and DJ, they talked about AI Canvas, and that really seemed to raise a lot of curious looks from the audience, as far as I can tell. Can you explain what AI Canvas is? I just worry that people aren't rocking it.
I can explain. It's a troubleshooting and execution workspace with multiple data sources. We wanted to bring in collaboration, and then we also wanted to bring in our deep network model, which is purpose-built, built with Cisco data, so that it's better than base models at actually solving problems. We're at the beginning of this journey, but let's jump into the demo, and I'll walk through some of the things you may have seen in the keynote and some other surprises that we have. Let's assume there's a ServiceNow ticket, and here's what you're seeing. It's pulling all of the information from ServiceNow and then talking with our deep network model to actually go in and auto-generate a dashboard of things that a tier one engineer might want to look at.
The tier one engineer can also add in an email to create a single source of truth, and then also can look at different things on the board. We want to be really transparent and share it with another user. She is going to share it with Will, generate a summary, and hand this off because it turns out it is a network issue, and Jackie, our tier one engineer, may not have permissions. When Will comes into the board, which you are going to see in a moment, I hope, he is going to have a completely different Canvas experience in terms of he is going to have the auto-generated UI, but his assistant is going to be different, and it is going to include the summary. He may decide that he wants to go in and look at some additional statistics and compare some application logs with availability.
We will auto-generate a graph, and he can decide what he wants to do here. He is going to add this to the Canvas, which is a single source of truth, and you can see he is using different data sources with Meraki and ThousandEyes. Lastly, he is going to go in and ask, what could I do to solve this? We are going to come back with our deep network model and talk about that there are three options. Keeping the human in the loop, he is going to decide to implement QoS, and it is going to do that. We are going to put another card on the board, understanding what was actually done, where Will also has the option to revert it. Because again, we want to be transparent, keep the human in the loop. Lastly, Will may decide, I want to monitor this.
In the keynote you saw, we had two cards. We have live cards as well as static cards. In this one, he's going to update the static card and bring in and see that he's actually solved the networking issue here, because you can see that his availability has the packets are no longer being dropped, and it appears that the network is back to normal. Lastly, he can generate a report. We heard from our customers, like sometimes sharing this out, making sure it doesn't happen again, is time-consuming. We auto-generate a report, and then he can pass that off to his leadership and his team, and they can analyze what happened and ensure it doesn't happen again.
That's amazing. The idea of even just generating a report for executive consumption, that is huge. I will say the part that I'm most excited about, by the way, is that it's not a separate product, right? It's like integrating into our network platforms deeply.
Yes, it's integrated into our network platforms.
Yeah. Hey, listen, maybe just final two seconds here. We obviously have a lot of exciting things happening at Cisco Live. Just quickly, what are you all most excited about?
I'm just stoked to be able to engage with all these customers and hear what use cases they are that can drive our roadmap. Our source of truth starts with you, so thanks for your participation.
You took my answer, but I'm excited too. I'm excited to hear the reactions and solve customer problems, because that's why we're all here to help you.
I'm excited because these are checkpoints for us. We announce things, and we launch them, and we get this instant feedback. That feedback loop is really important. I think hopefully you all feel, because we certainly feel that the vision is coming together, and our ability to do things as one Cisco, it puts us in a great position. To Nick's point, that doesn't happen without all the people in this room. It doesn't happen with the people that are willing to put the time in here with us. We really do appreciate it.
Yeah, and I think the most powerful thing is watching all of these different announcements that we're making and how the portfolio is coming together as one across security, networking, observability, and collaboration. That's, I think, a super powerful message and shows you kind of how Cisco can offer an innovation and portfolio that's great in some of the parts.
Fantastic. Hey, let's give this panel a round of applause. While they exit, I do want to bring up a special guest here. Jacob, if you could join me up on the stage here. Jacob here, he leads infrastructure for Novo Nordisk. We are going to have a short conversation with Jacob.
Thank you.
Please join me. All right. I don't think many of this audience know who you are, so maybe just a quick intro about who you are and a little bit about Novo Nordisk.
Yes, my name is Jakob Lilholt , and as you can see on my hair, I have also seen the internet involvement. We are a company, Novo Nordisk, based out of Denmark. We live in Copenhagen. If you do not know Novo Nordisk as a company, then maybe you know some of our products. We produce Wegovy and Ozempic. Over the last five years, that has at least brought the brand of Novo Nordisk to a lot more patients than we used to have in the past.
I didn't realize that. Maybe we could talk about discounts on products. All right. Hey, why don't we just start with the first question. You've been with us for a long time, really investing in Cisco to invest in your infrastructure, grow your infrastructure. How have you seen your business change and grow during this time? What are some of the network challenges that you've seen during your tenure there?
Yes, we have been with you for a long time, and thank you for the partnership. You said that we've been on the two case switch at some point back in the days, so I didn't even know that. We have had this growth, probably if you can imagine, that when you produce, and suddenly you get this product that everybody wants to have. Our biggest challenge in that sense was that we went from a production that runs five days a week, maybe 16 hours a day, onto 24/7, seven days a week, zero downtime, whatever that is. That has, of course, put some challenges on our infrastructure and on our network services that we need to overcome.
There have been challenges on the way, but it's a nice, fun journey to be on where you can really feel that if you build out the systems right, it's a very nice place to be.
Yeah, it's amazing seeing how you went from a five-day-a-week operation at 24 by 7. I'm curious, in this age of AI, it's a big topic at this conference and outside. How are you and your teams preparing for AI in your network, not only now, but in the future? How do you think about how AI can support innovation at Novo Nordisk and really the cutting-edge research that you all do?
Yes, it's true that our scientists, they love AI. For them, AI is getting, as somebody said, unlimited CPUs, unlimited bandwidth, unlimited access to all data. That's, of course, a challenge in our environment that we cannot provide that. We all know that even cloud is somebody else's data center somewhere. They were using cloud as the first place to run some of all these large language models and so on. Now we are really seeing them that they are also pushing us in the infrastructure departments to deliver some of these AI capabilities in the data center to train the models and run the models. That's a big challenge for us, right? First of all, to understand what is it that they're talking about when somebody said suddenly H100, and we know something about Cisco switches.
It has also been a learning for us to understand what is these new products and what is these new capabilities that we need to deliver. I'm pretty sure that we will deliver them because in the world of pharma, using AI models to train and figure out what the next molecule for us is going to be, that's a huge game, right? We have 15 years to market a product before it runs out of patents. That means that we would love to get a new one out every seven to ten years. Today, the research time is 20. It is very important for us to be able to use the AI capabilities.
That's great. I think understanding the business context obviously helps you and your team be much more effective at your jobs. That's what I'm hearing.
Yes, definitely. Of course, as you say, also in networking, we're very excited that something that was very large computers a couple of years ago now comes into a network where we can see and use the AI capabilities to troubleshoot. That's amazing.
Yeah, and you talked a lot about how Novo Nordisk is growing and expanding. How do you think what you and your team do with the network can help support that scale?
Standardization.
Standardization.
Standardization. Even though we have this era of AI, standardization is still the way for us to go. We can wake up any network engineer in the middle of the night and get him to troubleshoot something that he already knows. That makes it easier still. We have branches with the Meraki setups where there are two users assigned to. We have sales offices with maybe 50 or 100. We have large campuses with more than 1,000 users, so 5,000. We also have production facilities. All of them are somewhat equally important for the end user if they have a problem. Standardization across all that platform is extremely important for us.
Yeah, I hear you. I mean, it seems like I imagine that the branch of a handful of sellers to the manufacturing side to your corporate campus, standardization, they must. I guess you think of the touch points your team does to simplify the complexity. Can you touch a little bit upon that?
Yes, as I say, it starts for us by building these standard designs. They actually go also outside the portfolio of Cisco products because it's also standardizing across our out-of-band management, our UPS system, and everything else. We all know that when we are building something that is to a certain important level with a certain SLA attached to it, we know that we can deliver that as an IT organization.
Got it. The other thing we care deeply about, and I'm sure you do as well, is securing your infrastructure. Given this wide dynamic range of network infrastructure that you have to help secure from the compliance requirements, I'm sure are very prevalent in your industry. How do you think about that?
You're absolutely right. Let's start backwards because compliance in pharma is a must, right? We have FDA regulations, and we have the EU regulations, and whatever all these are called around the globe. Using AI to actually make sure that we can live up to these requirements is one of the things that we are investigating in a lot and making that simpler. The report you just showed on the AI Canvas, right? If I could get that on my compliance system, that would be fantastic. What are we doing to secure that? Of course, we are using a lot of the traditional things in terms of firewalling, but we are also very keen on OT security, like looking into what is happening in the OT network. That was not a thing you did five years ago, right?
It is definitely a thing on our CISO's radar today to understand what is going on in those environments and how can we keep those compliant and secure. Yeah.
I'm curious, and do you think about, I hear you may also look at things, observability tools as well, to help you with that. Where does that play a role here?
It does play a role because one of the important things about being compliant is also that you must all the time be able to prove that you are compliant, right? If somebody comes in and asks us, what is it that is going on somewhere? What change did happen here? If we cannot prove it to them instantly, they actually have the capability of shutting down our production, and that means that we can't sell any product in the end. That is, of course, the extreme case. That is what we are working towards in this pharma-regulated industry.
Hopefully, some of the integrations that we're doing with our portfolio with Splunk will be able to continue helping you with that as well.
Definitely. I can say I have a wish for you in terms of the AI Canvas, right? You did a change and implement it. I'm hoping it's documented in ServiceNow when you do it. That's our requirement.
All right. Lizzie, there's a feature request right now. All right. I love how this works here. All right. Hey, we're running short on time, but I'm curious to hear just the panel here. What are you most excited about? What do you want to go check out on the show floor after this?
Yeah, so.
Hi there, everybody. Welcome back to the Cisco TV studio. We have just wrapped the first of our Tuesday series of keynote deep dives. We are going to go directly into the next one. Now we're about to see a new approach to digital resilience. How can we get ahead of issues? How can we minimize business impact? We've got Cisco product marketing leaders Rajah Mukhopadhyay and Kevin LeBlanc here with us today, along with Michael Brand from Honeywell. Let's head straight out to that next keynote deep dive. Here we go.
President of Product Marketing at Splunk, Kevin LeBlanc.
Great. Thank you so much. In today's world, digital resilience is the ability to keep your organization securely up and running in the face of any disruption. Now, disruptions can include things like cyber attacks, bad code pushes, technology failures, or even network outages. Ideally, you want to prevent these things from happening in the first place by modernizing your infrastructure, keeping it up to date. Of course, you want to have the latest security measures in place. Even with all that, we know sometimes things just go bad. Being resilient is all about getting ahead of those issues and being able to respond quickly when and if they do occur. That seems like a daunting task, but you're not alone. We have pegged that problem. We've pegged the problem of digital resilience to be a $400 billion problem across the world's largest organizations.
Recently, we conducted some research on the cost of downtime with the Global 2000. What we learned is the direct cost of downtime is, for those companies, on average, $200 million per year per company. That seems like a lot. When you think about lost productivity, regulatory fines, and lost revenue, it all adds up. There is an even greater cost. It is the hidden cost of downtime. 94% of the respondents indicated downtime incidents have a negative effect on the velocity of continuing to innovate. What we are talking about here is a data problem. The ability to detect, investigate, and respond requires that you have access to the right data in the right hands at the right time. We all know across our organizations, data is typically in silos. The tools and systems that we use do not always talk to each other.
It makes it really, really hard to be able to be proactive in finding those issues before they cause business impact. We also know that the digital environment that we live in is growing increasingly complex. We have a tangled web of endpoints and infrastructure ranging from various flavors of public cloud, private cloud, hybrid, all the way to on-premise compute environments. Also, look at all of us here. We're all mobile. And what do we expect? We expect anytime, anywhere connectivity. When something goes wrong, we know what gets the blame. It's always the network's fault, right? Now, traditionally, the problem was relatively straightforward. Find what's broken and fix it. In today's world, it's more complex. You need to, yes, find what's broken amongst the multitude of things that are broken.
You need to prioritize what needs to be fixed and fixed quickly to avoid business impact. Now, AI. While AI presents an amazing wealth of possible opportunities for every one of us here, it also introduces new and unique complexities. AI brings opaque, nondiscriminate behavior, and that is challenging the traditional observability model. What that means is we need a new approach to how we monitor data pipelines, how we model performance, and we inference accuracy spanning our entire infrastructure, application, and our AI stacks. What that means is the problem is that when you do not know the business impact of problem A versus the business impact of problem B, you have to chase everything with the same degree of urgency.
That's really, really tough to do because you can't prioritize when you're not able to understand the visibility and context, which results in a lot of firefighting, which alerts a lot of wasted time, which has a direct result to revenue and can also have negative implications to customer experiences. What do we need to do about this? Only Cisco can give you the complete visibility and the context that you need to be able to do this by bringing together assurance, observability, and security. We are the only vendor that can provide the depth and breadth across data, analytics, and domain subject matter expertise to help you build greater resilience for your organization and also build awesome digital customer experiences.
While other tools do a pretty good job at red, green, yellow system health status, Splunk observability provides deep critical insights into the business impact of every problem. We excel at doing fast and focused troubleshooting. We've heard some things about resilience today in the keynote. We've talked about resilience here. Resilience is critical for business success. In some research that we've done across 1,800 IT professionals and engineering practitioners, those that have cracked the code on being more resilient are able to find and fix problems almost three times, 2.8 times faster than others. They can spend 38% more of their time, their precious time on innovation, and they've seen a 2.6x times annual return on their observability solutions that they use to make sure they meet their reliability goals and are able to deliver awesome digital customer experiences.
Over the past 12 months, we have taken Splunk AppDynamics and Splunk Observability Cloud and brought it into a single piece of solution. We are uniquely able to provide complete visibility across any environment in any stack spanning applications, infrastructure, and your AI environments. Also, what's more important is we've built it into a unified experience. While Splunk AppDynamics is the solution for hybrid third-party tier environments, we also have Splunk Observability Cloud, which is the solution for cloud-native environments. We leverage the power of the platform, the platform advantage, which brings in common experiences. We use the common data set, common UI, and all of this comes together to help you make faster insights and smarter insights. Also, this is all supported by Splunk ITSI, which gives you that bird's-eye view across your entire environment.
Only Cisco and Splunk are able to help you cut through all of that complexity. That allows us and you to help your executives, IT, security, networking, and security practitioners hone in on what matters most in a matter of seconds. Now, to help you understand more about how we get this done, I'd like to welcome Raja Mukhopadhyay , our Vice President of Product Management and Observability, onto the stage.
Thank you, Kevin. There are three primary reasons why customers choose Splunk. The first, we give them complete business visibility into any application, any technology stack. Second, we make it really easy for teams to find and fix issues in their environment and then tie those issues back to their impact on the business. Third, we make it really simple. We give teams a lot of control, a lot of visibility into their telemetry data so that they can better manage the costs associated with observability. Let's look into these in a bit more detail. With Splunk's offerings, we are able to bring together IT teams, engineering teams, security teams with a shared context around the entire environment in a way that no other solutions can. Now, we, of course, provide composite vertical visibility into individual applications: the front-end application, the server-side application, and the underlying infrastructure.
We do this across all application architectures and across any environment, all the way from a mainframe that may be running on-premises to a newer application running in Lambda on AWS. In addition to such vertical visibility into individual applications, we also provide horizontal visibility into higher-order business processes, business processes that cut across multiple different IT systems, multiple applications. For instance, with Splunk, a manufacturer can not only monitor their SAP environment, they can monitor the higher-order, the business process around the efficiency of their supply chain when the supply chain cuts across not just SAP, but can access another IT system. Similarly, if you look at a bank, they can not only monitor their core banking application, their mainframe application, but they can also monitor their credit card issuance business process.
They can get deep insights into where users are getting stumped in the sign-up flow, why getting cards out to new issuers is taking long, and such. Let's look at this in the context of one of our customers, TravelPort, how they are using these capabilities. They are a distribution technology and payment solutions provider to the travel and tourism industry. Prior to getting Splunk, their monitoring environment was fragmented. They had lots of monitoring tools. When they would have an issue, all of those different tools would be screaming alerts. They had upwards of 20,000 alerts per day, and they had zero context into what an individual alert meant in terms of its impact to the business. That is exactly what they were trying to solve for.
They wanted to bring the noise level around alerts down and to be able to better quantify what individual alerts' impact were to the business. That is precisely what they have achieved with Splunk. They have brought down the alert volume by 95%. They have done this by leveraging Splunk's capabilities to automatically group thousands of alerts into a much smaller number of human-relatable incidents, incidents that their operations teams can then chase down. In the process, they have brought down the mean time to detection by 75%. Not only that, now they have high-level dashboards that have summary roll-ups into their business-level KPIs. Things like call volume, things like the number of abandoned cards, things like their overall booking. Really deep visibility into the business, not just their IT systems.
Now, the second reason why a lot of customers choose Splunk is that we make it really simple for teams to find and fix issues in their environment. We do this primarily by giving teams a unified view of their full environment, a full-stack view, if you will, across logs, events, metrics, traces, so that users can quickly search and correlate across all of this data with ease. We also provide out-of-box monitoring dashboards for standard infrastructure components, so it's a really plug-and-play experience for users. In addition to automatically grouping alerts and reducing the alert noise, we also provide for automated anomaly detection. When engineers are troubleshooting because of an alert, we provide automated root cause analysis summaries, all to make sure that they are able to troubleshoot issues quickly.
Developers, they are able to inject tags, tags that inject their specific business context into the telemetry data, so that when they are troubleshooting, they can use those tags to do the troubleshooting analysis and be able to answer questions like, "Hey, this outage that we are having, does it impact all of my customers or only particular segments being impacted? If users are having a bad experience with your service, are users in all geographies impacted or only particular geographies?" Let's look into how one of our customers, Repay, they have been taking advantage of all these capabilities. They're a payment technology and solutions provider with an omnichannel platform.
They have billions of transactions that go through their platform, and the stakes of service availability are very high, both in terms of what it means for the business in terms of revenue, but also in terms of their regulatory obligations. Now, with the solution that they had prior to Splunk, it was patchy. Teams would take a long time to find and fix issues because the monitoring tool gave them only limited piecemeal visibility into their environment. They have switched over to Splunk, and now they have a comprehensive view of their overall environment in real time. In doing so, they have been able to bring down the total number of incidents as well as the amount of time it takes them to fix it. They have been one of the earliest adopters of our AI assistant capability.
Using that, they have been able to bring down the transaction, the amount of time it takes for them to triage incidents by 50%. The transaction latencies, they have been able to bring it down by 30% by optimizing their services. A lot of operational efficiencies that the team has gained using Splunk. Now, with the shift to public cloud and the shift to a microservices-based architecture, technology stacks have become more heavily instrumented. That has led to an explosion of telemetry data volume that comes out of digital environments. That is why we see a lot of customers struggling with the costs associated around observability. This is where Splunk, we help our customers stand up flexible telemetry data pipelines. We do that in a way that is open standards-based way with OpenTelemetry.
The great thing about OpenTelemetry is that you instrument your environment only once, and then you do not have to touch it, even though you retain the freedom to have multiple different observability solutions over time. The thing here that makes Splunk's solution unique is that we give our users a lot of control into the kind of processing a particular piece of logs data, metrics data, trace data will have, all with the goal of ensuring that the teams are always able to get to the most optimal observability posture that their business can afford. Let's look into how one of the customers, TravelPort, they have been getting deep visibility into their business in the face of massive data volumes. As you may know, Progressive, sorry, they are one of the largest insurance companies in the world.
They have upwards of 15 million web requests coming into their platform every day. It was crucial for them that they be able to zoom into any one of those requests and find out exactly what is going on in there. With their prior solution, they had a lot of blind spots. They did not know when users were having a frustrating experience with their digital offering and were falling off. That led to a significant loss of revenue. Coming over to Splunk with our full fidelity tracing solution, we now process upwards of 50 million spans per day for Progressive. That is upwards of 8 million traces.
What that means for the Progressive team is that they now have the flexibility that out of the millions of insurance codes that are floating around in their system, they can zoom in and they can find exactly what is going wrong with one individual code. Talk about finding a needle in a haystack. This has really increased the level of confidence that the business has around the digital offering. Lots of deep visibility into the business, even in the face of massive data volumes. These are some of the primary ways that our customers are getting value out of Splunk and why they keep choosing Splunk as a strategic partner that can help them improve the overall resilience of their digital environments. Now, let's get to the fun part. I'm really excited to share with you some of our latest product innovations.
Now, when users are having problems with a digital service, IT teams don't know a priori whether the problem is in the application, the underlying infrastructure, or in the network that connects the users to the application. This is because there is no single tool that spans the entire gamut and can present a summary view across the network and the application. This is where Splunk's IT service intelligence, ITSI, comes in. ITSI integrates with application monitoring tools, infrastructure monitoring tools, network monitoring tools, and is able to provide to teams a high-level visual map across your application and network so that when a problem happens, IT teams can very quickly go and rule out, "Look, we don't think this problem is the network.
Looks like it is the application, so that they can then engage the appropriate team so that the team can then use their specific domain-specific monitoring tool to troubleshoot further. Let's look at all of this live in a demo video. Could you roll the video, please?
Digital resilience starts with knowing, is it your system, your network, or somebody else's problem? With new integrations for Splunk ITSI and Cisco Catalyst in Meraki, as well as Splunk Observability Cloud with ThousandEyes, you don't guess, you know. In ITSI, you see the health of your entire environment in one glance. You see your network, Catalyst, Meraki, as well as your applications. Red lights indicate an issue. When there's a few, how do you know what's the symptom and where's the cause? Using the new content pack, ITSI pulls data from Catalyst in Meraki.
One click shows you critical services in Catalyst. One more takes us to episode review, where the new advanced AI feature in Splunk ITSI called EventIQ autocorrelates alerts. No manual rules needed. It flags store SJC12 as the root cause, and the events timeline reveals just how much noise EventIQ cut through, all without lifting a finger. Another click lands you with context in Catalyst Center, reviewing critical metrics. A key access switch is, in fact, down. Now you know it's the network edge and can focus on the fix. By the way, Meraki-managed branches also show up in your ITSI environment. We've got you covered regardless of what you use. If you're a ThousandEyes user, you're used to proving the network isn't at fault. Now, with Splunk APM integration, you don't just clear the network. You see what's actually wrong in the app.
When a test fails, like here in Oregon, you check the network path. It looks clean. The new APM service map shows up red. This links us to a corresponding trace in Splunk APM, which shows a 500 error in the app stack. You can pass that down to the dev. No back and forth, no waste of time. Whether you're a network engineer or app owner, Splunk helps you say, "Not it," or fix it. That's digital resilience with Cisco, now supercharged by Splunk.
That's pretty cool, right? As you saw in the video with ITSI's integration into both application infrastructure monitoring tools and the network monitoring tools, teams can now quickly identify when they have a problem. Is it the network? Is it the application? Is it the infrastructure?
Because the network is not always the problem, even though, as we all know, the network team always gets the blame. We already have available ITSI's integration with AppDynamics with Observability Cloud. We released the integration of ITSI with ThousandEyes a couple of months back. I am super excited to announce that we are extending that with ITSI's integrations into Catalyst Center and Meraki Controller. These will be available in September. We will not stop there. We will be extending that with integrations with Nexus ACI, as well as Cisco's SD-WAN offering, so that you have that visibility. When you know that the network is a problem, you are able to quickly isolate it to the subdomain. Is the problem on the WAN side? Is the problem on the campus side? Is the problem on the wireless wireline, the data center switching side, all of that?
Now, in addition to ITSI's integration with these products, as you saw in the video, we have done pairwise integration between Observability Cloud and ThousandEyes. The whole idea is that if there is a team that is using ThousandEyes synthetic stress to test their application, and if the test fails, right there in the ThousandEyes console, as you saw, the network team would get a summary view of the application. Again, the point is to let them know, "Look, is this the network? That's why the test is failing, or do you think it's the application?" If it is the application, with one click, the application owner can get into Observability Cloud, where they can pull up the traces corresponding to that failed test.
Now, as you heard earlier in the keynote this morning from Jeetu, a lot of applications are becoming AI-enabled, and newer applications are being stood up in an AI-native way. That has led to the emergence of an AI infrastructure stack for applications, all the way from the hardware with GPUs to the language models, to vector databases, to language model orchestration frameworks. We need to provide visibility into this new AI infrastructure stack for applications. Not just that, we need to provide visibility into the application code so that developers can look into individual prompts that are going into the models and the responses coming back so that they can make sense of it all. That is exactly what we are doing with Observability Cloud.
We will be providing out-of-the-box monitoring dashboards for all the way from NVIDIA GPUs to language models like Azure OpenAI, to Amazon Bedrock, Google Vertex AI, to vector databases like Vectara. We'll also be integrating with language model orchestration frameworks like LiteLLM, LangChain, OpenAI, so that developers can get the right hooks into their code so that they are able to look into the prompts and the responses coming back. We'll also give you an idea of the cost associated with the model. We'll look into things like the number of tokens you have used and give you a sense into the quality of responses coming back from the model, the percentage of responses that are hallucinations and such. We will be having all of these capabilities available in Observability Cloud by July end. For our AppDynamics customers, we'll have similar capabilities to monitor LLM-enabled applications as well.
Now, that is how we are thinking of providing observability for AI-enabled applications. The other side of the coin, if you will, is how we are using AI to really help our users in their observability practice. The key thing to realize there is that observability products, they're products of necessity. Our users, developers, SREs, IT operations teams members, no one really wants to be in an observability product. They're there because they have to. That key insight is what animates our broader vision of what AI can do for observability. We imagine a completely different product experience in observability powered by AI, where the human toil involved in running and maintaining digital environments is brought substantively down.
Whether it is detecting anomalies, whether it is troubleshooting, whether it is taking remedial action, or whether it is generating summary incident reports, we will be automating all of these flows with AI, with the human always being in the supervisory loop. That is exactly what you see us doing all across our portfolio. Within AppDynamics, we have always had anomaly detection for business transactions. We are going to be extending that to databases. Similarly, we have root cause analysis support for business transactions. We are bringing it down from the application layer into the infrastructure layer. For customers of AppDynamics that use the self-managed form factor, we would be introducing automated transaction diagnostics in the virtual appliance offering. Within ITSI, we have had this capability where you can group thousands of alerts into a smaller number of human-relatable incidents.
Now, in the past, in order to do that, users had to specify rules, rules that governed how different entities are related so that the software could group the individual alerts together. No longer. Now we'll have AI do that for you. That is the capability called EventIQ that we have released in preview a couple of months back. Finally, earlier this year, we have released the AI assistant in Observability Cloud, where developers' SREs can just ask questions in natural language and get quick responses around what's wrong in their environment, all with the goal so that they can get to the root cause faster.
Now, we will be extending that and directly embedding AI in the workflow so that when an alert goes off and the SRE team comes in, they will have right there a summary view of what software thinks, what the AI thinks has gone wrong, and what could be the root cause. A lot more stuff here to come from us. We have always been proponents that you need unified observability through your environment. What we mean by that is that you want a full stack view across the front-end application, across your server-side application, across the underlying infrastructure, and across logs, events, metrics, etc. Not only that, you want a view of the security profile of the application as well. All of this so that the human beings who are troubleshooting, they're not trying to look at individual pieces of a jigsaw puzzle.
They can look at the entire frame all in one go. That is why I'm very excited that we have brought together the best capabilities of AppDynamics and Observability Cloud into a single unified observability experience. Through that, you can cover your entire application estate, whether it's three-tier applications or whether it's microservices applications running on the public cloud. We have also integrated between AppDynamics and Observability Cloud so that if you have a three-tier application that you are using AppDynamics to monitor and it's running on Kubernetes, you can use Observability Cloud to monitor the Kubernetes infrastructure while still making sure that the developers that are in the AppDynamics console, they get a view of what's happening in the Kubernetes infrastructure right there within the AppDynamics console. Another capability that we are very excited is a secure application capability within Observability Cloud.
What this does, it gives engineering teams a view into all of the security vulnerabilities that exist in their code and from a runtime lens. We know a lot of engineering teams, they are overwhelmed with the number of security vulnerabilities that exist in the code. They need a prioritization framework, which ones to fix first. Getting that runtime view that these are the vulnerabilities that can be invoked in runtime is a crucial signal that will help teams get on top of their backlog around security vulnerabilities. Last but not least, we have the session replay capability in AppDynamics, where whether it's mobile sessions, whether it's browser sessions, developers, product managers, designers can get really granular insight into how users are interacting with the digital offering.
You can see lots of innovation that has come out, lots more to come in the next few months. We're going to keep continuing on that momentum. Of course, we have been recognized by the industry analyst community for the work that we do for our customers. Even more, it makes us incredibly proud to get the validation from our customers to see what we do for them and how we help them make their digital environments more resilient. To hear one such customer story, let me invite back on stage Kevin LeBlanc and Michael Bryan, leader for observability at Honeywell.
Welcome to the stage, Michael, and thank you for taking out some of your Cisco Live experience to be with us. Really appreciate it.
Happy to be here.
Awesome.
I'm wondering if you could just briefly tell us a little bit about Honeywell and the functions that you serve, and a bit about your team.
Yeah, Honeywell is a fairly massive corporation. We do a lot of stuff. For me, I'm specifically over the observability and monitoring section for application performance monitoring, so making sure that our applications are performing, staying up, and running correctly.
For such a large organization, how are you ensuring consistency across what you deliver for your internal customers?
It's all around setting up GDMs and SOPs, helping enforce adoption of those standard processes to ensure that we're actually having the processes followed correctly. We have our guidelines set up. We have what we need to happen all identified. We have to make sure that our teams are using those runbooks.
Perfect. Thank you.
Can you talk to us a little bit more about your main observability focus in terms of how you're using the Splunk tools today?
Yeah, so a lot of our focus is on synthetic browser testing. We like to simulate our customers accessing our sites. That way, we know what experiences they're getting. Above and beyond RUM, RUM is fantastic, but synthetic testing lets us identify those really intermittent issues that might be popping up, intermittent, like you just said, intermittently, yeah.
Can you share an example of how synthetic testing has helped you? Maybe describe how you're using that tools.
Yeah, so we use synthetic browser testing to simulate critical user journeys, test what our users are going out onto our sites, the specific key functionalities that we really require to be up and running in order for our e-commerce platforms to actu ally be functional.
. Great. Now, metrics.
Metrics are really, really important. Can you tell us a bit more about some of the most important metrics that you look at, investigate?
For us, it's mostly uptime and performance. We have a set of performance metrics that we've identified between business and DA. We've identified what is most important to our customers to ensure that our sites are functioning appropriately. We really look at how each individual function on our e-commerce site is. Is it up and running? Is it impacted? Is there degradation in the performance of it? We want to make sure that we see how our customers are experiencing our website.
Great. When you think about the journey that you've been on to help your internal customers, the external customers of Honeywell, what is some of the biggest value that Splunk has brought to you?
It's allowed us to get a single pane of glass, which allows us to troubleshoot faster. It allows us to consolidate our tools. It really lets us find the issues that are impacting our customers and then resolve them quickly and accurately. It's really all about the customer at the end of the day. We want to make sure that they have a great experience on our e-commerce platforms. That's what it allows us to do.
Great. Thank you. When you think about how has your team changed their day-to-day processes, their day-to-day life with Splunk being involved in the mix?
It's really allowed us to put a lot more focus on issues that we may not have seen in the past. It's allowed us to be more proactive rather than reactive.
We can identify potential issues that might be coming up and then actively work to resolve them before they have impact on our customers.
Great. You talked about this cultural shift from proactive and reactive. Can you tell me more about how that's impacted your team's ability to do what you do?
I think the easiest way to respond to that one is it's allowed our teams to put more focus on improving the platform stability of our product rather than trying to chase down a number of various issues on it that are reported from our customers. It's the idea of removing repeatable instances of failure and instead proactively improving the platform stability.
Great. In the keynote this morning, we heard a lot about detection investigation. We talked about it here as well. Can you tell us how Splunk has helped you get earlier detection investigation?
Yeah, so we use APM and synthetic tests to generate alerts that allow us to identify when we have failures occurring on our site. It also allows us to proactively identify when failures might be occurring in the future. It allows us to hop on those issues, quickly resolve them, and really bring a better customer experience to our customers.
Great, great. Can you describe something that you could catch now that before you just would have missed, would have never seen?
Yeah, so we had a PIF service, a backend APM service that had a node that was intermittently failing. It was something that only impacted a couple of customers here and there. With our synthetic test, we were able to see it in real time, iden tify where the node was, and then resolve that issue permanently.
Perfect. Business impact. We've talked about proactive, reactive.
Can you describe to me having a proactive state? How has that helped the business?
It's allowed us to ensure that we have fewer P1 and P2s. It's made our platform more stable. I mean, it's really increased the CSAT for our customers. It's provided a better experience. In doing so, we improve how it's perceived.
Perfect. When you interact with your executive stakeholders and leadership, what are sort of the metrics that you review with them?
Can you talk to us a little bit about how you do prioritization of alerts for critical path situations?
For us, we really put the most amount of focus on our actual critical user journeys. We have worked with our business teams to identify what are the core functionalities of the site that have to be up and running in order for us to have a successful site. Cart, Search, Checkout, those are all portions of this critical user journey. When we get alerts, we are actually only getting alerts on things that are impacting the ability for our e-commerce site to actually bring in revenue.
Perfect, perfect. How has synthetic testing evolved overall for you at Honeywell?
We went from bare bones tests, pretty much just uptime tests, to now doing a full critical user path.
We go step by step through everything that the user would touch, and we make sure that we are testing out each piece of that functionality.
Great. In the keynote this morning, G2 talked about it's not just about the network. It's not just about security. It's a heavy, heavy dose of collaboration. Can you speak to how you and your team collaborate across the technology individuals, across networking and security?
Yeah, we work with our BRM team in order to bring together everybody who needs to be brought together as soon as we identify that there's an issue occurring. What we found is that we get faster root cause analyses. We get fewer escalations because we're actually identifying the key members needed on these calls before the calls happen. It ends up being a cleaner BRM altogether.
Great.
BRM for the audience, what does that stand for? Business Resilience Management. Business Resilience Management, great. Can you describe some of maybe the metrics, performance? How do you engage? Is it dashboards that you share with this team?
Primarily, it's dashboards. We have a dashboard set up that shares the key critical metrics for business. We also have an IT operations dashboard, which is more of a command center desk level dashboard. It lets us know how the systems are working, whether there's performance degradation, and triggers alerts based off of predefined settings, thresholds.
Could you describe to us how observability has improved these cross-team dynamics?
I think it's allowed us to talk more freely to one another, to be honest. We aren't blaming one another for any specific thing. We already know where the issue is. We just have to find out how to fix it.
Perfect. Can you describe for us how you're bringing some of these challenges across insights to business? How are you showing up in front of your key stakeholders? Why not just visibility?
We've talked about the importance of visibility and context to business. For us, it's really about sharing the metrics that matter to them. We share the metrics. We were able to show them how the site's performing. We're able to identify those performance metrics and connect them with the CSAT scores that we're seeing on the business side to really show just the uptime and performance in a clear, quantifiable method.
Great. As you traverse from one key stakeholder group to the other, what are some of the main metrics that are resonating with your line of business, the business indivi duals?
It's all about performance. It's how fast is our website running?
Are the key functions up and running? What is our uptime status?
It's everything that drives our site's ability to bring in revenue.
This will likely vary from one company to the next.
What sort of cadence do you recommend with business alignment?
It's really going to be based off of the business that you're working with. We have multiple SVGs. We have some SVGs that we meet with weekly, some that are monthly. Usually, they're looking at these dashboards that we've created daily. I can't tell you the number of times I'm getting pings on teams going, "Hey, what's going on here?" It has really allowed them to see things in real time. While I would love to say that there's a set cadence of we meet every two weeks, they're very active in reviewing these data points. I think that it's helpful.
Great.
For the audience, SVG is what's definition?
It's one of those acronyms that I don't actually know. I think it's sub-business group.
Very good, very good, cool. Look, in the keynote today, we talked about it here as well. Is Honeywell using AI? Is your team using AI in the context of observability? What are you doing?
Yeah, Honeywell's gone all in on AI. We really believe in democratizing it. We've put AI in the hands of every single Honeywell worker. We're also utilizing Splunk's observability AI assistant, which has really helped us drive faster root cause analysis. We dive into fewer rabbit holes. We've actually been able to use it on occasion to build smarter dashboards.
Awesome, awesome. Why do you think AI is important to your operations?
I think it's important because it's really about shifting the focus onto resolving issues rather than trying to hunt them down and find them. The less time we have to spend finding where the issue is, the faster we can resolve it, and the happier our customers are going to be.
Perfect, perfect. We talked a lot about this, that observability, it's not just about tools. It's a practice. Can you tell us a bit more on where you're going with your journey and your practice?
Our next step is really going to be based around unifying our dashboards. I come from the e-commerce space in Honeywell. And we've recently expanded, and we're really pushing for full deep observability across all of our applications in Honeywell. What we're going to want to do is unify out our dashboards.
We want to be able to have a single pane of glass across all of Honeywell and then really start maturing the use of AI to resolve and find these issues.
Great. Maybe a narrow aperture, say in the next 12 months, what are some of the biggest growth areas or critical priorities you have?
It's going to be standardizing the use of observability across all of the applications that Honeywell has. That's going to be our biggest growth area. We're setting up the GDM and the SOP in order to really push and drive that because we don't want there to be separate silos of how people use observability. That's something we're changing.
Perfect. Michael, I really appreciate you spending time with us here this week. Thank you.
Thank you.
In closing, I'd like to leave with just a couple of things that we talked about today. We talked a lot about needing the visibility, but it's a lot more than just visibility. It's about visibility and context. That gives you the insights to be able to prioritize and take actions. Michael talked about trying to move from reactive to proactive. A reactive state is the only way that you're able to get ahead of issues to avoid business impact. One of the things that we truly believe is observability. It's much more than tools. Observability is a practice. Just like weights in a gym, if they sit on the corner and you never visit the weights, it doesn't do anything for you. It's about the practice. It's about the discipline. It's about the cadence of putting these things to work.
We have worked a lot with companies like Honeywell and Michael. We have learned and experienced what IT operations and developers and engineers are doing. I would like to leave you with a couple of things that we have seen that top teams that have cracked the code on resilience do time and time again. Visibility across the environment is critical. The visibility gives you the next piece in terms of the context. To get the business impact in context, you have to cut through that noise. You have to reduce the noise. We found that teams that are highly, highly successful in doing all this are leveraging AI, machine learning to get deep inspection and quick inspection understanding of root cause analysis. That is about being obsessed with minimizing downtime. The last one is collaboration. It is not just tools. It is not just an observability team.
It's not just the network team. It's how do you bring all this together through standardized practices, shared data, and shared workflows. If anything I've talked about today piques your interest to go deeper on whether it's observability, security, the combination of both in a DevSecOps world. I would encourage you to download and hit this QR code. There's a discount code to join us at .com in Boston in September. We would love to see you there. Thank you so much.
Fou r years ago, we wanted to change two things. We wanted to change the way we worked, and we wanted to change the sport of golf forever. The solution of both was Webex. My name is Cameron Pollard, and I play golf. When I first started competing, playing in front of people wasn't my thing. Being autistic, it was hard at the start.
The Webex Play Series came next. The idea was one tournament, one course, one title, one prize: purse, men, women, and all abilities together. Golf is a game that's hardly changed in hundreds of years, and it took some doing. Webex was showing things that we didn't know we could do in our sport, and it enabled us to connect our athletes all over the world to get our athletes ready for the global game. It's working for them. We've seen athletes such as Grace Kim start in the Webex Play Series to go on and win on the LPGA. I do want to show them that you're actually good and you belong there. I'm not just a bloke with autism. It's just because we're good enough to be there. It's a pretty awesome feeling. Since the inception of the Webex Play Series, female participation has grown 25%.
Participation in general over the last year has grown from 3.4 million to 3.8 million golfers in Australia. That is phenomenal. As for the team, Webex changed the way we worked, and Webex's AI has been amazing. People Focus is a great feature for us. It allows for everyone in the meeting to feel included and be seen and heard on the one screen. Real-time translation is very valuable for us at the PGA. We talk with global tours on a regular basis. Things used to take twice as long with our team out in the field. We began this journey thinking two things. We would change the way we work, and we would change the sport of golf. What we never imagined is that we'd actually be changing people's lives. Do it like this. Best customer experience is the one your customers never have to think about. Hi, Sam.
Sorry about your train cancellation. Let's get you booked on the next available train. Because you communicate proactively, anticipating their needs before they become needs. Because you deliver 24/7, always-on service with AI agents that offer personalized experiences. Because you can scale with agentic AI that completes tasks and resolves customer issues in the languages and channels customers prefer. Because you empower customer-facing teams with AI tools so they can stay engaged while providing higher-touch support. Because you harness AI to turn insights into action and enhance every part of the customer journey. Experiences so effortless and engaging, your customers keep coming back. That's the power of Webex and AI-first customer experience. Webex Calling Customer Assist enables your local and regional employees to deliver seamless customer assistance with AI-powered features in the collaboration app they already use.
Click to call allows your customers to get in touch with one click, no login required. Once they've reached out, any team member can answer and help. With AI Assistant, they'll have efficient tools at their fingertips, like real-time call summaries that can be shared when transferring a call. With the call sentiment for supervisors feature, supervisors can assist team members whenever they need support. Enable your local and regional employees to deliver seamless customer assistance empowered with AI in the collaboration app they already use. Webex Calling Customer Assist. Cisco and SAP have agreed on an ambitious milestone for the migration of our more than 250 offices globally. Right now, we are planning on having upwards of 350,000 devices on the network, including IoT devices. We have more than 100,000 employees.
They need best-in-class IT infrastructure to work wherever, whenever, and however they choose with the devices they prefer. With Cisco, we've made our network simpler and more secure. Cisco SDA also enabled us to improve our Zero Trust Access policy, ensuring only trusted devices can access corporate resources. With Cisco Spaces, we could leverage this foundation to establish a global platform for smart offices. Our first experience confirms our hopes and expectations that we can manage everything in one place together with a professional partner like Cisco. Tribeca has always been curious about technology. I feel that so much of Webex and Cisco specifically have done for us is help us lower the barrier of entry so that all artists can participate and be involved in the festival.
We're using Webex events in order to make the experience of De NiroCon much more interesting, immersive, and connected for our audience. A frictionless experience was key. We don't want our fans and customers worrying about how to find things or where to find things. All we want them to worry about is how do they have the best possible experience. We want to continue to tap into the broader suite of capabilities that Webex has, including a lot of the AI capabilities that they continue to build out. It's important to work with companies that have shared values because it's a partnership. It helps us amplify our voice. To a certain extent, it's a shared mission of Cisco to be able to amplify diverse storytelling. You need to have global partners if you're a global festival. We are truly honored and proud to be partnered with Cisco.
Meet AI receptionist for Webex Calling. Your always-on virtual receptionist that can answer questions, book appointments, and transfer calls so your business can boost customer satisfaction without adding staff. Thanks for calling Florissa Dental. I'm your AI receptionist. How can I help you today? Hi. Yes, I have a toothache, and I'd like to schedule an appointment. First, I want to make sure that you accept US insurance. Yes, we do. US insurance is covered 100%. Would you like to speak to an insurance specialist, or are you ready to schedule an appointment? I'd like to go ahead and schedule an appointment. Thank you. Great. I'll transfer you to a nurse who will give you an initial consultation and help to schedule your appointment. Thank you. Deliver outstanding customer service. AI receptionist, available later this year as an add-on to Webex Calling.
Take your business into the new era of customer experience with Webex AI agent, which answers questions and fulfills customer needs, all while delivering personalized conversational experiences. With Webex AI agent, you can engage your customers through the channels they prefer, using natural language and human inflections. Sounds like a special occasion. Okay. Based on your description and our stock, I would recommend daffodils. Oh, and I want a dozen, please. Understood. A dozen. Coming right up. Okay. I think we have what you're asking for. I've sent it directly to your phone. Even if your customer changes their mind mid-conversation. I've got it. Miami. That's the destination. Of course. I found several flights that fit your preference. With reliable guardrails in place, the conversation stays within the AI agent's capabilities. Could you also book me a taxi, by chance, for when I land? I'm sorry.
Unfortunately, that is outside of my scope. Anything else I can help you with? You can be sure that your customer's needs will be met, their problems resolved, and that all important transactions completed. Let's get that payment arranged. I've sent a text message with the confirmation. Have a great day. Welcome to the new era of customer experience with Webex AI Agent. Thank you for calling. You're caller number 25. Your estimated wait time is 37 minutes. Hi. I'm Blossom, an AI agent. Hey, how are you? For your rush delivery, I can help you find the flower you were looking for. We have— Yeah, it's my grandma's birthday, and I need her favorite flower, but I have no idea what it's called. It's bright yellow with, like, bell-shaped center, very thin stem. Sounds like a special occasion. Okay.
Based on your description and our stock, I would recommend daffodils. Oh, and I want a dozen, please. Understood. A dozen. Coming right up. Okay. I think we have what you're asking for. I've sent it directly to your phone. Yes.
Good morning and welcome. Thank you so much for being our guinea pigs for this new silent presentation. This is new for me as well. First and foremost, if you can hear me and you're excited to be here, the coolest, most powerful session at Cisco Live, raise your hand. We can hear? Okay. All good. All right. I have to tell you, a couple of years ago when we kicked off engineering sustainability across Cisco, I was trying to convince people why energy was such a big deal for this portfolio and for our work. I don't think I have to explain any of that today.
I'm not going to explain why it's such a big deal. What I am going to tell you is help you with some numbers so we can start to see and understand the magnitude and the urgency. Pick your poison, if it's the stick with compliance and regulations that are still happening around the world, or if it's the great big opportunity in front of all of us. One of the most recent studies I stumbled upon that I wanted to share with you is a McKinsey study that's forecasted between 2023 and 2030. The amount of data center energy consumption spend is going to threefold, going to a projected over 150 GW of power by 2030. Just so you know, for comparison, 1 GW is one nuclear power plant. Right? We're looking at forecasts of demand that our grid simply cannot support.
No change in how much data center spending is going. The change is how big can you dream? How large are those numbers? It's not just in the billions. They're talking trillions of dollars. Right? We're talking leaders around the world saying we're going to rerack and redo every data center in the world, and now we're building new ones. If you go over to the campus side, the regulatory pressure to look at energy buildings and bring in, from a real estate perspective, we still have a vacancy problem around the world of refreshing campuses, office buildings, centers to make sure that we have not only the power needed, but bringing renewable energy and the right experiences to those employees. You add on top of that a world shortage of skilled electricians.
Anyone who's done any work in a building will know the most expensive labor you're going to pay for is those skilled electricians. This is where we're going to focus. Where we have been focusing is around energy on both the networking side and the cooling solution. I have to tell you, this year, there are so many new exciting things that we have gotten across a milestone finish line to bring to you here at Cisco Live for the first time. Let's start to first understand across Cisco's entire portfolio, we've had the pleasure and the opportunity to build this foundation around energy management. We know on both the campus and the campus future-proof workspace side, as well as our digital resilience with the acquisition of Splunk, all of this data is going to need to pervasively come together.
What we've been doing very quietly behind the scenes is standardizing across an energy management platform. This energy management platform, we've taken five key metrics across our portfolio. We've standardized it. We've made it available for free in visibility. Across our major controllers, you can now see how much energy you're using, the greenhouse gas emission footprint, the carbon intensity, and the mix of that energy. By location, you can see how much of that energy is coming from renewable, not just any renewable. How much is coming from wind, solar, thermal? We look at the cost of that energy. We have taken our data and telemetry. We've overlaid it with electricity maps and IEA data. Now you have geotagging. The only people in the world who have access to this data are Cisco and the end customer.
As you think about how we can build platforms and services and where customers are going to begin, it's really important to understand credible measurement. You have energy management as a foundation. Then we look at, you can turn off all the light switches you want. You can try to go on standby mode as much as you want. That's not going to move the needle for our energy capacity needs. What's going to move the needle is new innovation. If you've seen maybe or heard some of this in the past, you may be familiar with the old current wars between AC and DC. You may have heard of fault-managed power. I've been talking about it for a couple of years now. If it's your first time learning about it, let me simply tell you, there is a new class of power called fault-managed power.
It was introduced into our national electrical code in 2023. It is what comes after PoE. Power over Ethernet tops out at 100 watts. When we introduce fault-managed power, Cisco is leading the charge in terms of IP. We are working with industry alliance groups to bring together partners in the industry and start to really educate what this new technology is. It is not only smarter and more secure. It is safe to touch up to 450 volts DC. I'm going to say that again for making sure that everyone's hearing that okay. It's not just arc-safe. It's not just ground-safe. It is touch-safe up to 450 volt DC. For the very first time at Cisco Live, we have a live demo for you with Panduit transmitter and receiver powering Cisco switches, powering PoE endpoints. It is also cheaper and faster.
When you think about the deployment, you no longer need certified electricians to do everything end-to-end. You might have low cabling labor to do the cabling and then maybe your certified electricians for that final fixture. This is a tremendous way to save on cost. It's more efficient, right? When you think about the end-to-end power loss of distance and delivery and line loss, all of that gets fixed because this is DC. Imagine, what would you do with touch-safe DC power up to 450 volt? Would you go to the data center? Would you go to the campus? Would you take it to EVs? The art of possible with FMP is something that we're still just starting with, and we encourage all of you to dream with us. That's the power side. What's new on the liquid cooling side?
For the first time at Cisco Live, we have a rear-door heat exchanger, direct-to-chip, and immersion, all on the back wall right over there. What's exciting about this is you can start anywhere on wherever a customer is on their journey, whether it's a basic refresh and they're trying to simply contain their air with hot and cold aisles and rear-door heat exchangers, or you want to start to introduce smart direct-to-chip. If you want to start sprinkling in immersion cooling, that is also possible in the same exact space. We're going to double-click a little bit into the data center because with some of these new innovations, we want to understand, well, just how effective is this? Right? When we think about data centers being such a large expense, electricity represents the largest expense in data centers today.
We know that rack consumption on average, data centers, legacy data centers are around 10 kilowatts a rack. That was a catch just to make sure everyone was still listening. 10 kilowatts a rack on average today, data centers. New data centers for AI are being architected starting at 100, 150. We're now seeing hyperscalers look at 400, 600, 800 volts to the rack. Who's going to service those? Right? We think about legacy. What does it mean when you have a legacy install and you're starting to add, like, bubblegum and sticky tape, all of these different systems working together? You no longer will be able to service the rack without a certified electrician wearing an arc flash suit. That's an arc flash suit.
If you've never seen one or wore one, think about going into a hot aisle wearing that and then hiring a certified electrician just to swap out a server. What we're trying to introduce is this concept around more integrated systems. An integrated rack design allows you to take those silos and remove the inefficiencies from the get-go. We're partnering with a variety of ecosystem vendors to make sure that we have validated architectures and reference designs to make it easier for customers to say, "Give me that block. We know it works and we know it's optimized." Right? The thermal cycling and the signal integrity that happens on those GPUs, which are the most expensive part of that server, will go away when you architect in systems that are designed to work together with maximum efficiency.
The concept of taking power and cooling and think about sometimes we're referring to it now as L11, taking wherever customers are in their current design and making sure we can transition them to the future. If there's a brand new data center build, of which there are many, how are you designing these new data centers with the maximum efficiency with power and cooling in mind? Hopefully some of this is starting to make sense. The thing that really gets me really excited is tracking and measuring it. As I said in the beginning, that foundation of energy management is where you have to start for credibility. We put numbers on this. We took a rack design that has 128 servers, and we designed what a legacy data center would look like, air-cooled. We found 65% of that power is non-useful power.
Can you imagine 65% of the cost that you're doing is going to just cool the systems or being lost by power, the distribution line loss, right? AC to DC, distance delivery, it's inefficient. Imagine combining all of this, putting it in an integrated rack system, throwing on fault-managed power, 380 volt DC, and getting down from 65% to 22% of non-useful power. Raise of hands. Who gets a little bit excited about over 75% energy cost savings in a data center? Yeah. This is just for a smaller data center. If you put these numbers to an AI data center, we've also run those numbers. Those costs for a legacy data center, energy costs for about a five to, let's say, like a 10 MW data center, $350 million an annual cost of energy. That's the far left column. Right?
We have these numbers and we can model them based on the designs. That's what's really exciting about these reference architecture designs. Now let's look at where has Cisco done this in the past? With routed optical networking, we combined layers in the network to bring more efficiency, better performance, less space. Right? This is not new to Cisco. Many of our hyperscalers and web scalers have adopted routed optical networking. What's very exciting about that is we continue to innovate across the entire portfolio. Routed optical networking has been around. We've seen tremendous savings in service providers. The numbers weren't even believable. We ran them, we ran them. I think Deutsche Telekom was citing over 90% in energy reduction bringing this technology because, again, they're converging layers of the network.
We've continued to re-architect and design, and we know that for AI-ready data centers, optical is very important. With Cisco's linear pluggable optics, LPO optics, it basically eliminates that need for the digital signal processing on both the transmitter and receiver side. Again, we're consolidating, consolidating, and simplifying the architecture. By doing that, we are seeing 50% less energy compared to retimed optics. That's 30% energy overall on the networking side with LPO. We have that to display and show you how all that works together on that back wall. When you compare and you start looking at legacy, where are we losing all of this power? Right?
The numbers seem sometimes too good to be true, but if you actually take into account and you follow the power all the way through, you can see that whether it's power factor loss, AC to DC conversion, loss in chassis just for the fans itself, a lot of these are lost in current data center metrics. Right? PoE does not measure all of that wasted energy inside of every one of those boxes. When you take a look at what we're trying to design and how we're working with engineering across the architectures to simplify from the get-go, there is a lot more that can be unlocked in terms of energy savings. For those who are trying to imagine, how in the world am I going to get FMP in this data center and what do those architectures look like?
There are a lot of ways we can start to do that. The first phase gets fault-managed power to the rack. Just removing those copper bus bars, removing copper bus savings material cost alone is huge. Right? What we hope to envision in the future is simplifying that with cable trays. And why is that possible? We see in the campus cable trays with PoE simplifying that architecture. Maybe it's the sidecars where you have an entire rack of power at the end of a rack. We are reimagining and we are looking and we are working with a lot of these new data center design builds for brand new data centers getting ready for AI. We are doing that working with an ecosystem of partners.
We have a new program called Engineering Alliances that you'll see peppered in a lot of our solutions behind you, working with key partners across storage, cooling, power, immersion, and non-traditional partners to help customers make it easier for reference architecture designs. We are going to go validate that third party and then make sure you have the technical briefs.
The technical solution guides to deploy much easier. When I think of all of the things that are happening and how you can sort of dream, how does all of this come together? What is an example of where all this comes together for one instance with the customer example? I had the opportunity to visit SoFi Stadium. For those who have not seen or visited SoFi Stadium, it's not too far from here in San Diego. We have a video that we'd love to share to walk you through a real customer deployment.
This story is not about buildings, but redefining how people build things. Big things, like reimagining the once-iconic Hollywood Park racetrack into a world-class destination for living, working, sports, and entertainment.
Architecture used to define spaces. Now it's equally about IT infrastructure. Partnering with Cisco, we've built a single, powerful, secure network that connects everything from broadcast systems to ticketing, security systems, and HVAC.
We have a converged network of intelligence with unprecedented control and visibility over our 298 acres. Since 2020, we've tapped into nearly one petabyte of data. We've solved problems even before they exist.
When guests visit and millions of people watch from home, the experience needs to be unforgettable. Everything behind the scenes must work flawlessly, and it does. This isn't just smart; this is revolutionary.
This story is about building revolutionary things, like SoFi Stadium, the most technologically advanced stadium on the planet. It's where 70,000 fans scream and stream at the same time.
At our big events, fans consume 32 TB of data without slowdown. Wi-Fi powers everything from the essentials to the innovative. It helped us improve navigation, reduce bottlenecks, and more.
The network empowers our broadcast team and our partners to seamlessly control and route thousands of video feeds throughout the Hollywood Park campus. This connectivity is truly transformative.
This story is about transformative things for the people behind the scenes.
With Cisco, we have the network and compute to scale operations and fan experiences. We're pivoting easily from intimate events to the world's most notable experiences. With full coverage through the network for both front and back of house, we could do anything for concerts, game days, and events. We're creating a dynamic environment while blocking 150,000 malicious threats annually.
The network has saved us millions, including reducing energy costs. We also cut our infrastructure data center footprint in half with Cisco. We can keep building.
This story is about building and building.
We started with SoFi Stadium, then YouTube Theater, followed by 900,000 sq ft of retail space and 25 acres of communal space. Expansion used to mean new networks and countless days of work. Now it takes hours, and everything is routed and ready. We keep building without missing a beat.
Or a championship.
With Cisco, we have a partner and a blueprint to build the future.
One of the things I love most about this example is Cisco's entire portfolio coming together and optimizing an experience. Whether you're a sports fan or an entertainment fan, I encourage you and invite you to go see this. If that's not enough, we have been in more and more getting ready for the Olympics at Hollywood Park. The idea that you can continue to build is something that we're seeing at campuses across the world, not only because of the data center moving closer and closer to the edge, but because we know that AI is going to make a huge impact on the requirements of where data and power need to go to every edge of the network.
These low deployments of DC and then looking at real estate and how all Cisco's portfolio can come together, we are building in and continuing to expand our ecosystem partners. What's exciting to me is when we see these projects, we're getting earlier and earlier in the process. The design in, the architecture firms, hey, there's going to be an Olympics here in LA. How do we build? Again, it's not a new network. It's just expanding on something existing. One of the things I would encourage everyone to think about is for all of the projects you're doing, how early in the process are you actually considering IT and the network? How has that changed from before, where IT is either an afterthought or IT gets built in after the building is built? That's no longer happening.
We're getting invited earlier and earlier in the process. There are so many partners in that ecosystem that we can do. When you take something like future-proof workspaces and you take PoE and you take FMP, for the first time again, I mentioned, first time ever at Cisco Live, we actually have fault-managed power with Panduit powering Cisco switches, powering PoE, endpoints and devices, all across the campus in our sensors and endpoints right behind you. Energy management is sitting underneath all of that. When you look at, well, what does that after effect look like in the campus? In our own Penn One offices with PoE only, we are saving 36% in our annual electricity costs, just PoE lighting. Panduit had a customer that was deploying three different buildings across 20 different remote locations.
What was fascinating is they saved 22% upfront in their capital costs and close to 80% in their operating costs for energy. Instead of weeks and months to deploy it, it took a matter of hours. When you start to compare all of the differences in how we build, how we design, and this idea of an end-to-end microgrid, DC end-to-end microgrid. For those who aren't aware, solar and wind are also DC native. 380-volt DC is where we already have an existing ecosystem today. As we start to think about deployments where we can put fault-managed power, 380-volt DC is effectively FMP ready. These ecosystem partners are who we are working with to design buildings and reference architecture designs.
If you know of a smart building or an intelligent building in your vicinity or area, I would go and ask, how converged or integrated are your IT and OT systems? If not, ask a question, is there a smart or intelligent building in your area that's connected to the grid and can do on and off loading? Do any of them have on-site generation? Do any of them use a DC microgrid today? These are questions that we're continuing to ask and continuing to design in across all of our customers. The Engineering Alliance program is one that I would encourage everyone. It's still fairly new, but what's exciting is the partners that we're partnering with to design in are seeing pipelines. We know the pipelines for building are longer, but the conversations that were happening are quite exciting.
The mega projects that we're seeing globally bringing in all of these various partners allow Cisco to build solutions and validated solutions with these partners and get them to market much faster. Some of the ways that we're continuing to do that and leverage Cisco's larger industry thought leadership is the industry groups where we participate, we lead, we're founding members and board members. When you think about the changing landscape and how rapidly the environment for regulatory is changing, standard is changing, the cost of labor, raise your hand if you're familiar with LEED certification for buildings. Raise your hand if you know something similar for data centers. ASHRAE and the US Buildings Green Council is looking at how to do something similar to LEED for data centers and standardize on that. That's not something Cisco is going to do alone.
We're going to work with these industry groups to start designing how we do things like that. That's going to help us all get a little bit faster deployment, safer deployment, and things that are better energy efficient-wise and better for the planet. On the right-hand side here, we actually have a consortium that we've come together in London. We're starting in the U.K. region. These partners are global, but it's a reference architecture design for what might be the most intelligent-ready building that we've seen and we've built, partnering with Schneider, Ideal, and Sadara.
When we take these architecture firms and we take these validated designs and we take something like Schneider's BMS and their systems and Cisco's architecture, it makes it that much easier for a hospital, a stadium, a school, a university, a new campus to say, let's leverage that reference architecture design and that ecosystem blueprint. There are a lot of themes that have come up. When I think about the innovation that we're bringing to you here at Cisco Live, and yet you're at the beginning of your Cisco Live week, I want you to take a look at some of these new innovations and ask the questions. For every project that you see, for every demo that you see, standardizing across power and cooling is a necessity. It's something that often we've taken for granted.
There are three key themes that I've sort of tracked and measured between the last time I was here and today. One of them is how we're getting designed in earlier and earlier. I mentioned in we're now talking to the MEP systems themselves. Think about the BlackRocks of the world for data centers, the Turners. I mean, there's so many and they're very local. But the manufacturing, electrical, and plumbing we're now talking about to make sure whether it's the primary loop in a data center or it's how to do on-site generation for a campus. Designing in early is a huge trend that is very different than just a couple of years ago. The second theme is around building these ecosystem of partners. There are so many cool innovations that we can bring faster to market if we can go validate those reference architecture designs.
If you happen to be in San Jose in the coming weeks or months, I would invite you to a new living lab that we're putting together, leveraging some of the great things you see here at Cisco Live in San Jose so we can actually have the technical conversations with customer design in. That third theme is around the thought leadership, like the industry groups, the alliances, the consortiums, where you start to see companies privatizing and coming together and trying to steer where we need the policy and the regulation and where we need private sector to be able to lead.
All three of those themes coming together are so important for the speed and scale at which power and cooling can no longer be taken as a given, but now need to be re-engineered, re-architected, and new technologies need to come a lot faster for adoption if we're going to have a hope and a prayer of getting this new technology responsibly built for the future. Cisco Live is a long week. You have a lot of choices. I am so glad that you've chosen to kick off center stage here at this session. I really look forward to conversations that are happening between the AI-ready data center space and the future-proof workspaces. Have a great week and thank you.
Welcome back, friends. Those were some powerful insights from Denise Lee. I tell you, on shaping the future of energy, I tell you, Cisco, we're not just talking about it, we're being about it. You know what I'm saying? First off, Cisco is leading the charge when it comes to innovation in this. When we talk about this discussion of energy, we are in it. We're not just participating in the conversation. We're actively transforming, hear me, guys, transforming how we think about energy, how we think about how we use energy in this digital world. First of all, I want to really express how I believe we're future-proofing is so essential and we're doing that here at Cisco. That means we're re-imagining how we are designing our energy grids. We are thinking about how we are designing our buildings, our spaces, our networks, our data centers.
We're thinking about that and we're thinking about how that's going to impact tomorrow's demands. Whether it's the future of AI is changing, we are thinking about that and we're making those changes right now, right here at Cisco. Finally, Cisco isn't just about thinking about long-term. We recognize the urgency. We heard the message of Jeetu and Chuck, that things are changing so rapidly. We're not just thinking about today, we're thinking about tomorrow. We're looking around corners. We're thinking about all those things that you may have questions about. I know you're here. You're coming to us because we are your trusted advisor. We're thinking about what you're thinking about even before you think about it. Say that three times. Bottom line, Cisco is more than just a tech leader. We're driving force and building a smarter and a more resilient energy future.
Now, Michelle, she had the opportunity to have a discussion with Denise when she came off stage. Michelle, we're going to look at that interview and see some more insights and see what's top of mind for Denise as she had that interview. Let's go over to Michelle right now.
Hey, everyone. I'm on the show floor with Denise Lee after her excellent speech. I was taking some notes. I loved what you said. I have a couple of questions that I want the viewers to see. What are the new trends driving changes in technology and building future-proofed systems in the campus and in AI centers?
Thanks so much, Michelle. It's great to be here in San Diego. Yeah, you're right. There are a couple of really big themes that have come out in recent months, especially with the explosion of AI. The first is bringing IT, traditionally an afterthought for buildings, earlier and earlier into the design process, all the way up with even the mechanical, electrical, plumbing, so MEP firms. We're talking about brand new data centers. We're talking about upgrading campuses. They're bringing Cisco earlier and into the process saying things like, where should we build based on access to power? How do we design the envelope of the building and the power requirements and the cooling requirements? That's just a brand new trend. Design in early is one trend. The other trend, second one, is around the ecosystems that we're thinking about.
There's a whole plethora of new ecosystem partners that traditionally Cisco would just assume we would go into and everything would plug in and work together. What we've learned is when you design these systems earlier in the process, you validate them, you have reference architecture designs, it just makes it that much easier for the customer to say, I want that solution and I know it all works together.
I love that. That makes a lot of sense. I'm so glad folks back home get to see this as well. All right. I heard you talking about energy management as a foundation. Can you elaborate on that and how are we helping customers integrate this into existing IT environments?
I'm so glad you asked that question because I have to tell you, it's not always the easiest thing to do, but it's the easiest thing to take for granted, which is you want access to good data that tells you how much energy am I using. You would think it's an easy question, right? And an easy answer, it really is not. There's actually a tremendous amount of work our engineering teams have done to standardize the measurement of energy across all of our devices and actually give it away for free in our platform. The visibility of this data is simplified, standardized, and free visibility across our entire platform.
I love that answer. Okay, one more for you. Strategy behind working with large ecosystems of partners to bring end-to-end solutions together. How do we do this?
We do this through a number of different programs. Anyone who's worked with Cisco will know our partner program is world-class. The number of partners and the kinds of partners we continue to bring in give us a fresh perspective, give us access to new markets and new technology and new innovation at speed and scale. It's really bringing our partners as part of the team and that extension. We just have to continue to make sure we're bringing in the right partners. There are a number of new ones, both in the campus and the data center. They're very different. For us to be able to offer our customers the best solutions, we need the best of breed vendors.
That's incredible. Thank you so much, Denise, for your time. Now back to the studio.
Wow. I don't know about you, but if I were you, I would be feeling a little FOMO right now because you're not here and you're missing it. I am here with my phenomenal co-host, Rob.
That's me. Yeah.
Rob, tell me, what have you been feeling? Let our audience know what they're missing out on because they're not here.
It is conversations like we just heard here. This is one thing I like about Cisco looking around the next corner because, yes, we want to put gear in all of these places. I think what there is on the data center side, there has been a huge understanding of, wow, these workloads are not traditional workloads. They have much different power requirements. The trickle effect of everything happens beyond that, which is how do we dissipate that heat? How do we get it? What I love what Cisco is doing is saying, there is even a bigger ecosystem that this needs to fit into because AI is pushing us into whole new areas, but it is also requiring whole new demands on infrastructure faster than our infrastructure traditionally has ever had to keep up with.
I'm glad that they're thinking about that because it's everything from liquid cooling at that kind of level to how do we coordinate across countries in such a way that we can also monitor, be responsible, and grade this stuff. I'm excited by the stuff we're doing. The demonstrations on the show floor are second to none. There are incredible things, many of which we featured on the show here. Yeah, being here, everyone I've talked to, it's really about the people that you interact with. Yes, the technology's there, but I love the people.
The people. You know I was going to just, I mean, you took the words right out of my mouth because I was going to connect back to the people. I feel like a broken record, but we can't do what we do without the people. They're the ones that's thinking of the things that others aren't thinking about. They're looking to us. I mean, I see the tea leaves. I can read the tea leaves. I mean, Cisco is looking up. I made that reference to the bat in the sky.
Oh, yeah, yeah.
I see the bridge in the sky. People are like, where is Cisco? We're showing up in ways that people can't even imagine. What are some of the, what's the one thing that you saw on the showroom floor that really stood out to you, Rob?
Oh, gosh. First of all, I don't feel like I've seen the whole showroom floor. Yeah, I've been a little bit focused on the areas that they wanted me to be halfway smart in. I’d focus on those things. That's been the NOC and the SOC. Just to stick there for a moment since it was what I was working on is fresh in my mind. I love the focus on they're using AI already in the network operations center. The reason why I love that is because it's not a demonstration of what a network could be.
It's a functioning network that is supporting what I would easily say is the pickiest, smartest audience you could possibly have to build a temporary network, which is something most engineers are smart enough to say, I don't know if I want to be on that project unless it's funded and staffed well enough. We do it on a regular basis, like that, consistently improving it. The NOC and the SOC, the teams that support that, that's still one of my favorites.
Have you had the opportunity to notice the looks on people's faces? Have you saw some wows or some come-to-Jesus moments, to use my reference, where the light bulb is actually going off and people are like, wow? Are you able to see some expressions from some people's faces?
I feel like that element is there. I feel like it's more on my face and maybe I'm not able to see theirs because I have this imposter syndrome where I figured that I'm constantly surrounded by people who know more than I do. I'm like, that was a big surprise to me. I'm like, should I let them know or do I need to hide that? I think it's happening at all different levels. Maybe some of us are holding it back to a certain extent. I love the move from generative AI to agentic AI because it's really hard to stay ahead of this curve right now. It's running hard on all of us.
I feel like we've got a lot of movements in the right place: hardware, software, control plane, management, and the way we're integrating clouds and the tools that have been announced. I think those are good wow moments.
Wow, wow. Lauren, I know she's going to share with us. Lauren, over to you.
Hi, everyone. It's Lauren, and I am here in the showcase area talking all things digital resilience. Let me tell you, being resilient in an IT world is not easy. Cisco and Splunk are better together and showing you that it really could be. The three things we want to think about when we think about resiliency and how hard it is are, A, there are more and more attacks. The surface is expanding. We have a number of different pieces and areas and mobile workforces that makes it really tough to secure everything. There is the downtime that we see with applications and how costly it is for businesses, running up to $200 million a year in some cases. It is like, what's the root cause of it all? Do not forget about how AI just really accelerates all of those things.
What we're doing here at Cisco Splunk is looking at it into three components. A, let's look at how we can power the SOC of the future and really start finding a unified way to see threats, find the meantime to resolution quickly, and understand what happened quickly. We're also looking at applications. How can we start being proactive and understanding unplanned downtime before it comes up? That third piece is all about the network. How can we have more of a way to see if there is a problem, where that problem is within the network? There are a lot of cool things that you'll learn about here. Check us out, Digital Resilience, here in the showcase. We will be here all week.
Wow, there are some exciting things, special things that are going on in the showcase. I know Michelle, she's out there. She's seeing some phenomenal things as well. Michelle, what are you seeing out there on the showcase floor?
Zee, thanks so much. The showcase floor is exciting as always. I'm here in the Purpose Pavilion. Right here is the Sustainability Showcase. There are a couple of things that I'd love to show the audience at home. If you're here, please, this is a call. Come check this stuff out. It's awesome. One thing I want to start talking about is the circular design score that we use at Cisco when designing products. Every product that goes to market is built, has basically scored on a 100-point scale, and they need 75 points to bring it to market. You may be asking, how do you get to 75 points? You start at 100, and points are reduced and taken away if they do something, if they build something that isn't sustainable.
For example, if there's paint on it, if it's made with something that's not recyclable, those things may decrease your score. We have competitions for each of these teams designing things. I think that's what keeps things fun, and everybody wins. What I want to point you to right here is the smart switch that G2 introduced and announced this morning. It is Energy Star certified, reducing plastic usage, and is up to 96% efficiency. This is incredible and showcases Cisco's commitment to sustainability. Speaking of goals, Cisco has had three 2025 goals that we have achieved, the first one being elimination of single-use plastics. This is huge. We've now also reduced the use of foam and packaging down to, or we've reduced it by 75%. See this old foam packaging over there? That is something we don't use anymore.
We actually end up using these recyclable packages right over here. This is made out of cardboard and much better for the environment. The last thing I want to talk about is the take-back program. If you come over here, look behind me. Pat, you can see this. They are actually participating in the take-back program. Anything with the Cisco logo is provided either a second life or can be recycled. This is just another great example of our commitment here at Cisco to sustainability. Zee, I want to take it back to you now. This was just an awesome opportunity. We'll see you on the show floor. Now back to you, Zee.
Thanks, Michelle. That is super amazing. Rob, what were some of the things that you received from that?
One of the things is I'm always amazed. There are so many things that go into being a hardware company on a global scale and all the different things that go into it. Packaging, for some reason, has always fascinated me in the sense that because it's very customized packaging to wrap that stuff up, ship it, and all that. You care about getting the product there. No, you've also got to think about, well, how are we going to recycle this? What's going to happen afterwards? Our customers receive, depending on if they're staging it or if our partners are working with them on this, that stuff's got to go somewhere. We would like to put it back in the system so we're not drawing upon, of course, raw resources to do that.
I like the commitment to sustainability that we also show here in the interactions that that drives for people on the floor. I do encourage people to go by and check those things out because it's a hands-on activity, which is always a little bit more fun.
Absolutely, absolutely. We are invoking our customers, those that are here, to participate with us. We want to partner with people that share our core values. That is important to us. I love that about working here at Cisco. What are some other things that are top of mind for you or that really pulled on your heartstrings as you were listening to that from Michelle?
From that is, one, just Michelle's doing such a great job. You guys being new hosts and such here, and it's such a pleasure to work with both of you. I keep thinking I'm going back to what we're doing with AI and some of the stuff that's been announced and the fact that we've been building our own security models and the way in which they're each, every demonstration on the show floor and the stuff that I've come across has got, there's certainly a base level of integration, but we're getting better at understanding how do we make sure that the right data is in the right place and only the right data so that we can prevent hallucinations and different things that can send us in a different direction.
Because we, like every one of our customers, is not only struggling to say, not struggling is not the right word, but striving to say, how can we provide this for you? Also, how can we provide it for ourselves? Because we all are going to move a little bit faster than we're comfortable for the next great while, I think. We have got the tools and the people to do so.
You know, Rob, double down on that hallucinations. Break that down a little bit for me a little bit more.
One of the challenges when it comes to generative AI, is what I'm speaking of specifically here is generative AI is unique in the sense that it's, and this is why it's scary and something needs to be understood from any other technology revolution we've ever had. It's the fact that it creates something from nothing. Or not nothing, but it creates something from whatever you give it. It's trained on who knows what, depending on what's been done. You have frontier models and such that we see publicly like ChatGPT, Gemini, many, many others. Then you have what we're seeing now, which is companies are saying, this is the important assets of my company. I'm going to feed this in. The training has to include some additional kind of worldly intelligence.
The trick is that these GPTs, the way they're built, they are designed when they generate answers to always give you an answer. And darn it, they're not a confident answer. You really have to say, you're a smooth-talking friend, but I really need to recheck what you're giving me before I go put my neck out on it, as we're seeing with attorneys. I think multiple have gotten in trouble for calling on cases that never existed. They're dealing with volumes of text. You've got to be really careful with these technologies.
Yeah, absolutely. You've got to fact-check the fact-checkers.
Yeah.
You've got to do that in this day and age.
A little bit you can do with the LLMs. Ask it again, and you'd be surprised how it can improve upon itself when questioned. It's actually very polite about saying, hey, I think you missed something, and it apologizes. I'm like, don't feel bad, which is probably unnecessary. I'm like, we start making this emotional connection with this inanimate object, and it's a challenge.
I do too. I'm always please and thank you.
Yeah.
Please and thank you. Can you please?
Why not? Because when the machines turn against us, not that that's the score that any, and that's not really happening.
Somebody's taking notes.
I just want to say he was at least nice. You know?
Perfect, perfect. I love it. I could sit here with you all day, Rob, and just yap, yap away because it's such great content. There's so much that we could talk about and so much that we're going to say, so much that we're hearing right now. We're about to step into a deep dive. We're about to go back to center stage. I want you to really lean in because this is going to be something critical that's really going to, I think it's going to really change the trajectory of what you were thinking before you got here, not just what we heard in the keynote. We're going to hear so much more in this next deep dive. I want you to get ready to dive into what we have next. You're about to listen to these other people that's about to enter the stage.
Look, let's go to the main stage right now. I'll quit talking. Let's go.
Things and maybe agents work together in the future. I want to start with, I think, what Jeetu started at main stage, the proliferation of agents. You look at this picture and you're thinking next year maybe an agent, some sort of autonomous agent is making this presentation, right? I'm going to paint a different picture, a picture that happened six, seven months ago when I went to and they suspected appendicitis. Did an MRI scan. You know where you wear this gown where you tie in the back? I wore that gown, and it is the MRI scan, and I'm lying there for 30 minutes waiting for the doctor to come.
It's at that precise moment my brain thinks, what if the agent on my phone could talk to the agent on that MRI machine, do a quick digital pathology, and tell me very quickly if I have something wrong or not, rather than waiting for the doctor? That's the kind of agent we are talking about, agents that make things productive in everyday life, not just robots. The network was not built for that. To do what I just said, you need a gig of throughput, 20 millisecond latency for uncompressed 8K video to transmit between these agents. We haven't built the network for that, have we? With that, look at the security risks with what I just described, privacy risks, right? What if there's a rogue agent? These are very sensitive environments. We haven't built it.
Guess who's going to get all these questions about the network, about security? Any guesses? It's all of you. There's a profound skill shortage among IT and networking folks. You are a rare species, ladies and gentlemen. Not going to have more of you, but you're going to do more work. That is why we had to require a network that's purpose-built from the ground up for AI. That is why I think you need a new architecture for this AI-ready secure network. We're looking at three pillars: scalable devices ready for AI. I know G2 had operational simplicity. I flipped it because, of course, I love hardware. You're going to start with hardware. Security fused into the network and operational simplicity for AI. A new era of networking begins today.
I love the video that Jeetu played, so let's go ahead and roll the video again.
Introducing a new lineup for the AI era. Smart switches to scale networks simply and securely. Secure routers to protect data and devices across thousands of sites. Intelligent Wi-Fi 7 to delight even the largest crowds. Cloud-native campus gateways for seamless roaming in the campus. Wi-Fi and URWB for ultra-reliable zero-loss access everywhere. Advanced firewalls for distributed threat detection. New rugged switches to power AI-ready factories. The secure network for AI-ready campus, branch, and industrial IoT.
That's cool. The video is cool, but I like steel. How about you all? I want to call to stage the man of steel, Lee Peterson, VP of Product Management for Audio. Let's do a steel unveil and look at these devices, Lee.
What's under here? Should we open it?
Yeah, let's go, right? Here you are, the brand new switches, wireless access points, and the routers. Sorry.
Let's go.
One second. I thought, Lee, this is this lineup of smart switches, secure routers, campus gateway, Wi-Fi. We've never refreshed everything together through the campus and branch, have we?
Never seen this much more than once. It's amazing.
This much lunch. The complete portfolio reinvented from the ground up. I thought, let's take a few minutes to go through each of these a little bit in more detail. Give me two minutes for the switches, will you?
Please. Only two?
Yeah, yeah, just maybe more than two. The Cisco 9000 Series, what can I say? Bekashi would say it's the beauty and the beast. The beauty can transmit a billion packets per second, and the beast, four billion packets per second. Bring on ludicrous mode. Bring on digital pathology. Bring on remote factory. We've got you covered. We are building these switches for the future, right? A billion packets per second is fast, but not smart. What is smart is that figuring out among those billion packets, the few that have security issues and directing those to a co-processor. What is smart is figuring out that the billion packets per second can go to zero in a second to save energy and hibernate and wake up in a second.
What is smart is that billion packets per second, an AI small language model right there in the switch to filter out what data to actually send to Splunk or the management tool so you can troubleshoot these switches. I mean, all of this could sound really futuristically, and you might be like, this is great. I'm going to wait a couple of years before I see the robotic agents. This is what I'll tell you, and all of you are thinking like that. I'm going to quote Paul McCartney and what he said to Julian Lennon, "Hey, dude, don't be afraid. You've gone. You're made to go on and get her. The minute you bring her in the network, things will start to get better." The Cisco 9000 series, made from the ground up to make your network simply better.
The smart behind the smart is powered by Silicon One, high-scale latency, internet-scale deep buffering without compromising on the bandwidth, flexible and programmable ASIC, and power performance, right? This thing, if I were to quote a Beatles song again, I'd be like, here comes the Silicon One. I say it's all right. I see the smiles returning to your faces. I could keep talking about the switches and Silicon One and Beatles all day long. Lee, why don't you talk a little bit about the routers and the APs as well?
Sure. No Beatles lyrics, I'm afraid. A little quicker. I'm going to start with the 8000 Series Secure Router. I'm a little bit biased. This is my product. I have to warn you. This is our 8200 Series Secure Router. Just to set everyone's expectations, they do not come with LED lights and clear like that. My son was very impressed when he saw this, but I had to assure him that's not how we actually ship them. That's just for demonstration purposes. There are a number of elements in this that when we think about what we need in the branch—sorry, where are we on moving around the slides? When we think about what we need in the branch, we hear consistently that throughput is of a high concern, not just as the demand for throughput, but the availability of throughput.
I tell people this to much jealousy. My house in Silicon Valley has a 10 gigabit per second symmetrical fiber running to it. That's a residential connection. The speed available at each of the locations now is increasing, as is the demand and the use for them. Greg mentioned the 8K uncompressed video use case. That's becoming more and more real. The ability to do more at the edge from an AI perspective is driving the demand for throughput, and we need to have devices that are built with that in mind. The way we achieve that with this device is through the secure networking processor. That is essentially using Cisco intellectual property to accelerate cryptographic workloads. That has a number of benefits.
One of those is being able to reach these higher order speeds, be able to run firewall services at near close to line rate as we possibly can on these devices, as well as being ready for when quantum computing, quantum cryptography becomes real. We are at Cisco working on this ourselves. We need to have a stronger security posture in the devices, and that has to be done at the hardware level. The secure network processor is the piece that enables all of that. The main thing for us is that the secure element of this is also around not just firewall services, which we can run very well on this, but also how we take cloud security. How do we integrate with SASE? How do we integrate with the components in terms of cloud security? That is baked right into this product as well.
Next, we'll talk about the campus gateway, which unfortunately we don't have up here. This has been a big ask for the longest time, where we want to take the mix of the benefit of an on-prem deployment with a controller and the scale and the efficiency that can be achieved with that, but the ease of use and the ease of deployment that comes from the cloud. The campus gateway is really how we mirror both of those things together, being able to achieve that scale locally, being able to do those things that are best done at the edge. Let's pick a large, we're talking about healthcare here a lot. A large campus environment where I'm going to have thousands of users, if not tens of thousands, and the ability to do layer two roaming at the edge.
This device allows you to do that with a campus gateway, but then also be able to have that simplicity of the cloud deployment, cloud management. It is a great solution because it can be deployed brownfield. You do not have to throw out your existing cabling or your existing devices. It can be dropped in and used very easily. That is available today. That is actually shipping and available today. The final thing is that within our white so Greg, they told me not to pick this up. Do you think I should try? You did say it was the man of steel.
The man of steel.
It's not that heavy. Come on, guys. That is completing our wireless lineup. This is our stadium Wi-Fi. This is the first time we've purpose-built an access point for stadium environments. I was talking to the CEO of the Dallas Cowboys, and he can't wait to get these things deployed in AT&T Stadium. These are fantastic because they're purpose-built for those types of environments, incredibly demanding, lots of devices, the expectations of the users in those levels, and the ability to use this to drive consumer experiences in those sort of locations is huge. We now have a full lineup of Wi-Fi 7 products here available, and I'm very excited about to round it out with the stadium products.
Thank you so much, Lee. I'll call you back on stage. Let's talk a little bit about security fused into the network, right? What are we doing? I'm going to focus a little on the campus, and Lee will come back and talk a little bit more about the branch. The first thing is it's sort of a layered approach where there's comprehensive security fused into every layer of the network. We start with the device itself, so the switch, the router, or the wireless LAN controller, if you will. What we've always done is secure boot-up, right? We make sure that you can't tamper with the Cisco devices. Now what we're doing with all these next-gen devices is making it quantum-proof. Even if you use quantum computing, you can't hack in and tamper with that.
That's the first level of security. The second is securing the network. We did line rate IPSec, line rate MACsec. This morning, you heard G2 speak about LiveProtect. I'm going to double-click into what are we doing with LiveProtect to protect the network traffic. The third layer is obviously the zero trust, right? Least privileged access, users, endpoints, and app, all the way from IoT device or your phone or your laptop, all the way to the app. What are we doing to do security, right? That's this. At every layer, we're going to have network play a key part. First, what I want to talk about, which is net new, is this LiveProtect. The concept is simple. I'm not going to do a demo, but I'm going to talk through what you're seeing here. The concept is simple.
If you have a PSIRT vulnerability, right? Last two years, we had 20 of those. Not all of you upgrade immediately, do you? You're waiting for a couple of months. Unfortunately, state actors, like in the Salt Typhoon attack, know this, and that's the precise moment they attack your network. What we're now doing with the Tetragon agent, the eBPF technology we have with Isovalent, we're embedding it into the kernel of the IO6C code and seeing if it can compensate for those PSIRT vulnerabilities till you get to upgrade, right? Of the 20 PSIRTs in the last two years, 15 of those we could compensate. What we are seeing here is a future view. It could be in Meraki or Catalyst Center, where you have all the PSIRTs, and it's first telling you which PSIRTs are more relevant for you. That's point one.
Given your config, is this something you should upgrade? It says we don't presume you want something in the kernel of the code, right? It could destabilize. It says, hey, do you want to put it in observation mode? This is let's put it in. Let's see if it can observe if that vulnerability is being attacked. Then it's observation. You can see like, yep, and then you can actually deploy it. All of this is possible. We're thinking through, given it's something that could destabilize the network, thinking through how we do this gradually. That's one. As regard what's net new in the branch with SASE, I think Lee's going to go deeper into it.
What we would like to do is bring in our best-in-class SD-WAN with secure access that's getting amazing reception together to create a secure branch with SASE, a unified branch that Lee will come and talk about where we can see, protect, and perform without compromising performance everywhere, right? We have zero trust in the branch. For zero trust in the campus, what's net new is you heard Jeetu talk about these switches being HyperShield compatible, right? We are doing that first in the data center. What this is, is the switches have powerful CPU. I spoke about it, a co-processor that can host the HyperShield agent. That agent basically is integrated with Cisco Security Cloud Control. The same policies that you put in here can be applied. Initially, we'll start with L3, L4, right?
Hopefully go on to L7 so that you're getting to a point where you have every switch port become potentially like a mini firewall, right? You have this distributed hybrid mesh firewall capability that Jeetu and Tom spoke about that's there in the campus environment as well. That's what's happening with security fused in the network. Above all, I think operational simplicity. We spoke about that skill shortage as I kicked off. That's probably the most important thing we need to nail to help you get ready for this AI-first world. We are starting with something that you asked about a lot, bringing together what was a Catalyst full stack, Catalyst hardware, Catalyst Center, and the Meraki full stack now with this next generation of switches, APs, and routers, common hardware. The same hardware can be used.
It's a Cisco switch, a Cisco access point can be used for both Catalyst and Meraki with a common license and a unified management. This is cloud on-prem. We'll meet you where you are. We are not coming up with yet another management tool. I want to be clear. It's just we're bringing together Catalyst Center and Meraki in a way in which you can consume it from where you are. I'll talk a little bit more about that. What we are doing, for example, and especially in switching, this is a big leap from where we were a year ago. Cloud can now manage all of the Catalyst fixed access switches, right? Shortly, your Cat 9K is completely managed in Meraki dashboard. We have an end-to-end assurance. I see Joe sitting here. ThousandEyes integrated with owned and unowned infrastructure.
We have that. We are bringing to life AI Assistant. Rather than talk about it, I think I'm going to invite someone special to do a demo. When you think of artificial intelligence, you want to bring someone who's naturally intelligent, Vimesh, to talk a little bit about our management demo.
Thank you, Greg. Thank you, everyone. Now that we have unveiled our switches, our wireless access point, the routers, let's bring up the HQ hospital. What I'm going to do is here the installer is going to go start racking and stacking the switches. The switches are going to come up. He's going to get a Cloud ID. As they get a Cloud ID, they're going to add the Cloud ID into the dashboard. This will add the device into the inventory.
Now you can go and add that device to the network. What we are bringing is a flexibility to our customers that you can add a device in a way where the configuration is managed to the cloud. In some cases, you already have the set up. You want to keep the business continuity there. We are bringing the capability where the configuration is managed from the device itself. Here I can onboard and bring the device with that hybrid mode where the device configuration is there. Now I'm going to add the username and password for that. The device will then connect to the cloud where it'll start sending the configuration, start sending the telemetry, and will be onboarded to the network. Now that channel is set up, you will start seeing that it is onboarded. I can see that 9200 CX is onboarded.
I see the cloud connectivity set up. I can see the cloud port status. The other capability that we have added is Cloud CLI. This unlocks all those IOS XE enterprise capabilities to be used by the network admin from our dashboard. You can log into the terminal and start configuring any of those enterprise capabilities that we have on the IOS XE platform. Here I've just made a host name change, but you can make any of those configuration changes. Now that we have brought up the device, let's actually see how can we secure and segment our network. With those enterprise capabilities, now we are introducing Fabric to our dashboard. This is where we bring those segmented networks to our customers through operational simplicity. Here I'm going to be setting up a Fabric. You will have a workflow where you can go and start configuring the Fabric.
I'm setting a San Jose Fabric. That is the HQ that I'm bringing up. I'll select the devices that need to be participating, selecting the role. The new thing that we are bringing in our dashboard is the ability to stage configuration. As the network admin is configuring, they can look at what devices that need to participate, what should be the role, the VRF that should be there, the border configuration for that external connectivity to be there. They can go and deploy. Now those configurations are being pushed down to the devices, and they come up. Right there in the same place, you are looking at the health metric. It just simplifies the entire lifecycle where you are pushing down the configuration and looking at the health.
I can look at the entire topology of the spine, the border, the health of the VXLAN tunnel. This is how we are redefining assurance. On the topic of redefining assurance, let's see now how with ThousandEyes, we continue to do that. We don't just tell you if the application is down, but why the application is down. More often than not, that why is hidden in the network. ThousandEyes has the power to pinpoint those network-related issues deep into the network, whether it is your public infrastructure or third-party infrastructure that you don't own. Here, ThousandEyes has the ability to pinpoint those. Here, let's take an example. An application was down. Here I'm seeing a 500 internal server error. We don't just tell you the symptom. We actually tell you the entire story. When I click on this service map, this is where things get interesting.
We are bringing data from Splunk Observability Cloud that we are telling the entire trace from the client to the network, to the front-end application, to the back-end services. This is distributed tracing end-to-end. Wait, there is more. I can look at the availability down correlated when the issue went down. I'm going to pause here for a second. What you see, this operational event, this is where things get interesting. It is not network pathing anymore. This is cloud infrastructure intelligence. Here, Cloud Insights is actually correlating the application down with the cloud configuration changes. I can go down to look at my back-end service where the change was made. I can look at the entire configuration change. This is end-to-end assurance. I'll see that when I'll go back to the service, you will see that it is completely correlated.
When the application went down, that is when the configuration change happened here. The service went from healthy to unhealthy. This is end-to-end assurance. We know that it's not always just the application. It is also the part of the owned network where the problem happens. This is where we bring some of the wireless troubleshooting capabilities. More often than not, you are constantly told that the wireless is not working. The client is not able to connect. What we are bringing is intelligent packet capture that actually captures the packet the minute the client is experiencing issues. Here, we are not just putting intelligence in the software. We have deep hardware ring buffers that are capturing those packets the minute the client is experiencing the problem.
As that client is roaming across different APs, each of those APs are capturing those packets and stitching it together to tell the entire story. We give it to the AI to give the outcome. Here you can see that it is giving in a description. All that it has done, it has just fed to the AI. It is giving a simple description that a client was attempting to connect to the wireless network using a WPA3. Initially, the authentication and reassociation was successful. The four-way handshake, you know, timed out, and the client tried again. AP temporarily rejected it. The way it is able to understand the AI model, because it knows at every packet transaction, it is looking at every packet transaction, and it is reasoning with it.
As it is reasoning, you know, many times you tell us that I have a skill shortage. How do I train? We are giving the entire chain of thought that how it came to that conclusion for each of those packets, how it reasoned with it so that your junior engineers can again be trained and get the right skills. What you see here in the PCAP viewer, in the interim of time, I'm not showing that, but you can actually go down and look at the individual packet, which packet it was looking at. Here, AI is going to highlight and tell you that was the packet that resulted in that conclusion. We are bringing entire visibility. This is not packet capture. This is packet capture on steroids. This is end-to-end visibility correlated end-to-end assurance.
This is client experience, network path from owned and unknown, secure segmented network. This is the power of unified platform. There is more to come. Greg, it's time to talk about more.
Thank you, Vimesh. Thank you. That was awesome. The question that gets to mind as Vimesh is presenting is, what if you're not naturally intelligent like Vimesh, but you still want to troubleshoot as well as you could? That's where the AI Canvas is coming to life, right? First is the new networking skills for AI Assistant. We'll start there. You can just natural language ask it the same questions that Vimesh was asking. We're going to present it because I want to go and talk about AI Canvas a little bit. You saw in the keynote demo.
Let's do a collaborative demo, if you will, someone who is artificially challenged, natural challenged, and naturally intelligent person. I'm going to start the troubleshooting with AI Canvas. Let's go to the demo. What we have here is the Meraki dashboard. I'm looking at it. It's an Assurance AUG summary view.
Wait, wait, wait. One second. It's there. It's not flipped.
Yeah, let's flip to the.
Laptop.
Laptop, please.
Maybe. Wait. Is it like you have to?
Yeah. OK. Here we go. In my case, I'm not going to go and ask about packet capture. I'm like, dude, I have a ticket. This is a ticket number. People are complaining about issues. What should I do? That's what I've typed in here. I click it. It's like, all right, let me look at it for details. This is AI Assistant in Meraki. It pulls up and it says, yeah, it looks like five folks are having issues. It is in Building 23, floor three. It's my floor, floor where I sit. That's where they have issues. It says, I can tell you more about these issues if you can open AI Canvas. Of course, I'm going to say yes, Vimesh, is that OK? I say yes. Let's go ahead and open it. Here we go.
Now this gets into this is part of the Meraki experience. It's not a separate product. It takes you to AI Canvas, which automatically pulls all the data relevant to that ticket. What it has done here is pull data from ThousandEyes. That's the first one. It's looking at the whole path and telling that there's a problem with LAN. Switches never have problems, Vimesh. This is Joe Vicaro's hack. Some of the switch seems to have a problem. It pulls data from Meraki in that same chart correlating. This is dynamically generated, showing looks like that there is a peak utilization, right? Like that's going on. I feel smart already, right? It says, hey, you can resolve this issue by doing a QoS change.
This is where I'm like, remember, I'm not a double CCIE like you are, Vimesh. I'm like, should I do this QoS change or not? I'm like, hey, wait a minute. Tell me more, right? Could you please provide more details on this? I do that. It's investigating. It's saying, pulling up more information again from various tools. You'll see. Let's give it a minute here. What it's doing is it's saying, OK, these are the five clients. Here is what's happening. These five clients are trying to do a server upgrade from looks like Windows. That's clogging the network. It's correlating the Webex issue that I saw. It's at that exact moment when the switch utilization is high. It's not the switch after all. It's the folks in the third floor doing server updates, right?
We found that out. If I now understand why I need to do QoS, it's saying you should prioritize Webex traffic if that's what you want. Pretty cool. I know if I do something, my IT team, I'm not allowed, right? My last name ends with AI. That means stop at some point, right? This is a point I want to collaborate. This is what's, I think, so super interesting where I'm like, OK, this is good. I'm going to call in Vimesh here, but not call in physically. I'm going to go in and say like.
He's going to come in. Yeah.
OK. That's what's happening. I'm going to write a message. The great thing is normally when I call Vimesh, I'd be like, dude, something's up. Please help. This time it can be smarter because all of the diagnostics I did, I can paste and generate an AI summary. This is automatically generated. It says this is what's going on. I invite Vimesh to look at it. Vimesh, come on in now.
Yeah. Just like what you saw. What happens is in many of the environment, you have the service help desk, right? They are the first line of defense. They are troubleshooting. This is where AI Canvas, they are going to be doing the first troubleshooting part. The same view that Greg, the service help desk, was troubleshooting, now me, Vimesh, the expert, is able to see and analyze to make that that is the right decision or not. Now I can see that, yes, it is right. I can go ahead and confirm. Let me go and apply the QoS policy on this. Here I log in. I'll look at that entire view. It gives me an entire summary.
I can validate all those changes, look at the data, and take that recommendation that the AI gave and go ahead and apply on the network.
How do you know this change worked, Vimesh?
Because it is monitoring. It continuously monitors the change. It tells me that it'll take 15 minutes. I can go back and again look at what was the change after the QoS policies was applied. Here it'll again give me a graph. I can see that. I can place it as a documentation itself and then log it in my entire SOP of how we came to a problem. The service technician did the troubleshooting and handed off to the network admin to make that change.
All right. Thank you so much, Vimesh. That was just wonderful. I think hopefully we brought to life how we are changing the nature of the game in the management tool of your choice. Here it was Meraki. Hopefully soon it'll come to Catalyst. We're thinking of how we can have this be a unified experience. We talked a lot about campus. Let's flip to the OK, it's there. We can also manage hundreds of clinics and remote sites. I'd like to invite back to the stage my man of steel, Lee Petersen, to talk about the branch.
Good. I did that deliberately. Yeah. Just to prove that things go wrong. It's not all perfect. Yes, we're trying to manage at scale. We focus a lot on the branch locations. Why they're important for us is that, and why they're important for you as customers is that these are unique in their nature. There are many of them. There could be tens, hundreds, thousands, tens of thousands of them. They're distributed. You typically have no IT staff there. They're super critical because this is where your customers have their impression of you. If you have a bad experience in a branch, we don't really want to walk into a, let's say, a retail store. We don't really want to walk into a health care clinic. The systems are down. It means I can't be checked in.
It means I can't do point of sale transactions. These things matter. What we're very focused on is not just the hardware, but also how we wrap a platform around that that's able to help us get to where we need to. One of the consistent challenges we hear is not just the scale and the distribution of these sites, but just the complexity of deploying and managing them. You can have the best plans in the world about what I want my configuration in the branch to look like. Actually keeping that consistent and keeping that over time to be the relevant thing is very hard to do. This is really what we're looking at around how do we make this easier? How do we make this better?
Now, the beautiful thing about when we talk about a platform, all those things that Vimesh showed, all those great benefits Greg talked about, all the amazing things that we get with Assurance, ThousandEyes, AI Canvas, when we take a platform approach, automatically all these products get that benefit. A lot of the effort has been around making sure that wireless, switching, and routing are all manageable by a single consistent platform. We saw some of that there. That then gives us a layer where we can apply intelligence consistently so we can do troubleshooting across all these assets, but also with AI Canvas across other domains as well. Where we've been spending a lot of our effort, though, is around how do I actually take one of these and deploy it in that consistent way? How do I do it quickly?
How do I do it accurately? How do I do it in a way that makes sure I'm not exposed from a security perspective? Talk about the platform already. What we really want to be able to do is think of that as like a programmable asset. Just show of hands, who's heard of Terraform? Probably maybe 20% of the room. Terraform is a really amazing tool where we can look at how do we actually do infrastructure as code. It's very common, used a lot in data centers. It's open source. It's an automation framework for deploying and configuring infrastructure. What it really allows us to do is to take what is today, Vimesh showed you the process of clicking through and adding a device. You can do that for a few devices. You can do that for a handful of devices.
What if I'm deploying hundreds of locations in a night? How do I do that consistently without there being error? That's really the big advantage we have with Terraform, the ability to treat the configuration as code and be able to have that pushed down to the device and be managed in a very consistent way. I'm going to show an actual case of this. Ignore the eye chart there. It's actually much simpler than it looks there with all those boxes. I'm actually going to talk about a specific case here. Let's imagine a scenario. I've got three active sites. Let's say clinics, right? Because we're talking a lot about health care here. I need to deploy five more. What would that actually look like? How would I do that?
Like I mentioned, the idea is that I want to be able to store my network configuration the same way I store code, right? Having it expressed in code in the way that it is here. We're using GitLab here to be able to store the code. I now want to go and make a change to be able to deploy five new locations. I'm able to use the same tools I use for developing software and deploying software. I'm able to go and use those here as well. Inherent with that is the certain things that you do. You don't just deploy software without configuring it and verifying it. What we're actually doing here is pushing a configuration to the pipeline around adding new branches. I go back over to GitLab here.
You can see I can actually check in on that task as it runs. Again, because I'm managing it the same way, my CI/CD, continual integration, continual deployment pipeline, I get all the same verification and checks I would. I probably wouldn't allow a low-level operator to push a change at this scale without configuration or without validation. We are able to use the benefits again of both the pipeline from a deployment perspective and Terraform to verify and validate those changes are good before they get deployed out. We can see here that that's completed. That now gets reflected. This is the logical deployment back in the Meraki dashboard that shows that that configuration is completed.
What that would allow me to do is that those five branch locations now, rather than having to fly a network engineer on site and have them do all the configuration on the fly, these are now logically deployed. When you get to the branch, each one of these pieces of equipment, it's a plug-and-play experience. It connects. It talks to the cloud. It gets its configuration. It's up and running, verified, validated, and working. That's how we really reduce the vectors of attack from a perspective of having a config drift, having security holes that I forgot to add a policy, things like that. It's a very consistent way to manage at scale and to manage with the high level of accuracy and efficiency that we need to. That's enough from us product managers.
I would love to actually hear from one of our customers, one of our amazing customers here. I am very happy to be able to welcome Aruna Ravichandran, our SVP of Marketing, to the stage, and Greg from Duke Health. Sorry, Jeff. Jeff.
I'm going to sit on this.
Thanks.
OK. Very excited to have Jeff with us from Duke Health. I'm Aruna Ravichandran. I lead marketing for the collaboration as well as the enterprise networking part of the portfolio. Jeff, give us a little bit of more color about Duke Health. Tell us about your role in the company as well as your organization's mission.
Sure. I've been with Duke Health for 17 years. Duke Health is one of 25 academic medical centers in the United States. Currently, we have four hospitals and about 160 clinics across the state. Our mission is advancing health care together.
Awesome. Let's get started. Tell us a little bit about your journey into Cisco. What were the challenges did you actually have with Duke Health?
Yeah, sure. We've got old buildings. We've got brownfield environments with legacy equipment. We are lucky to have a couple of greenfield buildings that we're getting ready to do. We're putting in the new 9350s. We're planning for the Wi-Fi 7 as well with those.
Even before you got started on the journey, I know that you are an EFT customer for these amazing smart switches, which we recently launched. Give us a little bit more color in terms of what were the challenges you were actually experiencing and why did you actually bring Cisco?
Again, it's the structured cabling that's in the buildings that we have to go and replace because of the speeds and the feeds for the new technology require that. Those are definitely a challenge.
Awesome. Everybody is talking about AI. You have heard about AI in the conference as well. Tell us what are some of the interesting use cases you have actually seen? I think you shared this with me last week when we talked in terms of how AI is being used at Duke Health in order to basically promote health.
Sure. I'm excited about this. I worked on an application called Autism and Beyond, which uses a smartphone to look at a toddler's facial expressions and tics to determine if they're autistic or not. What that does, it allows that toddler to get treated earlier for autism. Another AI-infused application that I worked on was called SepsisWatch. If you're not aware of what sepsis is, when your body goes into sepsis, your organs start to shut down and the mortality rate when that happens is very high. It's a bad thing. What we did is we took all the data from the vital signs monitors for a patient and used AI to take a look at that data to determine if a patient was going to go into sepsis before they actually went into sepsis.
Which by doing that, we were able to give the doctor a five-hour head start before their patient even started to see symptoms of being sepsis. By doing that, we found that we've been able to save about eight lives per month.
Wow. Wow. That's amazing. You know, these kind of AI advancements actually require a lot of infrastructure support, right? Give us a little bit of more color in terms of how Cisco has played an important role in terms of bringing these kind of AI use cases to market.
Yeah. So we've got medical telemetry that we use to allow for cardiac patients. What that does is now that cardiac patient who had open heart surgery, let's say, they need to get up and around and mobile to reduce their recovery time. By being able to do the medical streaming and have a device on that patient, we're now able to have that patient reduce their recovery time. Another thing that we do is we use Cisco Spaces as a middleware for our, what I'll call, a hybrid RTLS environment. What I mean by that is we've got our traditional RTLS environment, but then we augment that with our wireless to get RTLS across the entire enterprise and not just in inpatient rooms. That really helps to create clinical efficiencies. We also use it for staff duress.
You were talking about the other AI use cases. I know that you are an early EFT customer for the 9350, which we have just launched. Can you give us a little bit of more color in terms of how your experience has been?
Sure. We are fortunate to be part of the early field trials for a couple of products, the 9350 as well as the Wi-Fi 7 APs. What that allows us to do is to not only test and kick the tires of the technology, it allows us, more importantly, to be able to get ready for a seamless migration of this new technology into our architecture.
Awesome. Tell us a little bit more in terms of what do you think is going to be next for Duke Health? How do you think you're going to see Cisco continue to help you in terms of your needs and promoting better health for the community?
Sure. Today, Duke Health serves about 1 in 25 North Carolinians. We want to expand that to about 1 in 10 North Carolinians as we grow. By doing that, that means more patients, more staff, more locations. We need the technologies, the innovations, and more importantly, the Cisco partnership to help us reach and achieve those missions and goals and visions that we're looking for.
Awesome. I want to see whether the audience has any kind of questions we want to take. We have a couple of minutes left. Anybody's got any other incremental questions they want to ask Jeff at this time?
Come on. Somebody's got to have something.
Can I borrow the microphone and I can bring it to him? Do we have another one?
I think you said at the top that you're, I think, your infrastructure, you're building a net new hospital and really getting that to state of the art and retrofitting some of the others. How do you kind of balance looking at infrastructure when you're looking at how do you bring something up to code while you're bringing state of the art?
I mean, that's really a good question. It's this dance that we have to do. It really gets back to, I think, lifecycle management and being able to budget and plan for these things to come up. Because we just don't have the budget to do everything at once. We have to understand what is the priority for us and then budget for that and then have this lifecycle management that allows us to continue to bring these new technologies into our health care system.
I'll ask you another question. You saw the AI Canvas, right? You also saw the demo, which Vimesh as well as Greg did. How do you think AI Canvas as well as the AI Assistant is, what kind of a role do you think it'll play? Will it help you simplify your operations? Will it help your team actually manage everything in a single console so that you can get rid of those troubleshooting tickets? Where do you see the value add with the AI Canvas?
I'm really excited about the AI Canvas. When I saw that this morning at the keynote speech, my mind just started swirling around saying, oh, we can use it for here in the service desk to help the core network ops folks work with it. Then, oh, it's multi-gaming. I can have other people chime in. The core ops folks can go out and reach out to an Analyst 1 or an Analyst 2 if they need to for support. It's all real time. It captures all the data and everything. I see it helping us to really get to a point where we're reducing our mean time to resolution.
That was the ultimate goal. When you think about the whole agentic ops lineup, which we launched, when you think about AI ops, it's basically probably going from a couple of hours down to a couple of minutes. With the whole thing, with this entire lineup, which we have launched with AI Canvas as well as the AI Assistant, it's about mean time through resolution gets down to seconds. That is the ultimate goal with insight with people like yourself and your team. One more question. Yeah.
Hey, how is your security posture evolving due to the embrace of AI at Duke Health?
As you can imagine, as a health care institution, security is at a premium for us because of HIPAA and PCI that we deal with. As we go on this AI journey and look to incorporate that more in the day-to-day functionality, I see that it's going to be critical for us to validate that. It's going to be this journey that I think we go on where as we get more comfortable as we validate, trust but verify kind of thing, validate it, then we'll see that, oh, OK, we can really use it for security. It'll reduce a lot of times when things are happening, some kind of security breach, we can just let it go and do it instead of, hey, it detected something.
Now we've got to look at it and make sure it's doing what we think it's supposed to be doing and then let it do it.
In addition to that, especially adding on to what you said, with the AI Canvas, you'll now have an opportunity to share that with your security counterpart.
Absolutely.
Right? An opportunity to, again, accelerate the troubleshooting.
Yes.
Awesome. Please give a huge round of applause to Jeff from Duke Health.
Thank you.
Thank you for joining us today. That brings us to the tail end of this particular session. You can learn a lot more across the board. The next session is going to be around the unified branch, which actually starts in the next 30 minutes. If you want to participate in AI user experience research, please scan the QR code, which is on my left-hand side. You will have an opportunity to actually win our Bang & Olufsen headset, which we have in collaboration with Cisco. Thank you for joining us today.
Meet AI receptionist for Webex Calling. Your always-on virtual receptionists that can answer questions, book appointments, and transfer calls so your business can boost customer satisfaction without adding staff.
Thanks for calling Florissa Dental. I'm your AI receptionist. How can I help you today?
Hi. Yes, I have a toothache, and I'd like to schedule an appointment. First, I want to make sure that you accept US insurance.
Yes, we do. U.S. insurance is covered 100%. Would you like to speak to an insurance specialist, or are you ready to schedule an appointment?
I'd like to go ahead and schedule an appointment. Thank you.
Great. I'll transfer you to a nurse who will give you an initial consultation and help to schedule your appointment.
Thank you.
Deliver outstanding customer service. AI receptionist, available later this year as an add-on to Webex Calling. Take your business into the new era of customer experience with Webex AI Agent, which answers questions and fulfills customer needs, all while delivering personalized conversational experiences. With Webex AI Agent, you can engage your customers through the channels they prefer, using natural language and human inflections.
Sounds like a special occasion. OK. Based on your description and our stock, I would recommend daffodils.
Oh, I want a dozen, please.
Understood. A dozen. Coming right up. OK. I think we have what you're asking for. I've sent it directly to your phone.
Even if your customer changes their mind mid-conversation.
I've got it. Miami. That's the destination.
Of course. I found several flights that fit your preference.
With reliable guardrails in place, the conversation stays within the AI Agent's capabilities.
Could you also book me a taxi, by chance, for when I land?
I'm sorry. Unfortunately, that is outside of my scope. Anything else I can help you with?
You can be sure that your customer's needs will be met, their problems resolved, and that all important transaction completed.
Let's get that payment arranged. I've sent a text message with the confirmation. Have a great day.
Welcome to the new era of customer experience with Webex AI Agent.
Thank you for calling. You're caller number 25. Your estimated wait time is 37 minutes.
Hi. I'm Blossom, an AI Agent.
Hey, how are you?
For your rush delivery, I can help you find the flower you were looking for. We have.
Yeah. It's my grandma's birthday, and I need her favorite flower, but I have no idea what it's called. It's bright yellow with, like, bell-shaped center, very thin stem.
Sounds like a special occasion. OK. Based on your description and our stock, I would recommend daffodils.
Oh, and I want a dozen, please.
Understood. A dozen. Coming right up. OK. I think we have what you're asking for. I've sent it directly to your phone.
Yes. The daffodils are perfect. Can I order a bottle of champagne as well?
That sounds delightful. Unfortunately, we don't offer alcohol purchases online, only in store. However, we do have artisan vases. Would you like to add one to your order?
That sounds good.
You got it. If you're ready to complete your purchase, I can text you the payment link.
Yes, please.
Let's get that payment arranged. I've sent a text message with the confirmation. Have a great day. Hey, Elizabeth. Still there?
OK. Change everything.
No problem. Where are we going now?
The perfect getaway for my sister and I. Skiing, wine tasting, spa.
Sure. Sounds like a great time. I'll get right on that. I found a number of vacation packages available in your time frame. What day would you like to?
Wait. I've got it. Miami. That's the destination.
Of course. I found several flights that fit your preference. Window, eighth row, left aisle.
You're amazing.
Thank you. I try to be my.
Actually, what about Greece?
Introducing Webex AI Agent.
What's my account balance?
Is this covered by my employee benefits?
I'd like to exchange this for a different size, paint color, and style.
Where hold times end and self-service begins.
Four years ago, we wanted to change two things. We wanted to change the way we worked, and we wanted to change the sport of golf forever. The solution of both was Webex.
My name is Cameron Pollard, and I play golf. When I first started competing, playing in front of people wasn't my thing. Being autistic, it was hard at the start. Then came the Webex plays series.
The idea was one tournament, one course, one title, one prize curse. Men, women, and all abilities together. Golf has hardly changed in hundreds of years, and it took some doing. Webex was showing things that we did not know we could do in our sport, and it enabled us to connect our athletes all over the world to get our athletes ready for the global game. It is working for them. We have seen athletes such as Grace Kim start in the Webex plays series to go on and win on the LPGA.
I do want to show them that you're actually good and you belong there. I'm not just a bloke with autism. It's just because we're good enough to be there. It's a pretty awesome feeling.
Since the inception of the Webex plays series, female participation has grown 25%. Participation in general over the last year has grown from 3.4 million to 3.8 million golfers in Australia. That is phenomenal. As for the team, Webex changed the way we worked, and Webex's AI has been amazing. People focus is a great feature for us. It allows for everyone in the meeting to feel included and be seen and heard on the one screen. Real-time translation is very valuable for us at the PGA. We talk with global tours on a regular basis. Things used to take twice as long with our team out in the field. We began this journey thinking two things. We would change the way we work, and we would change the sport of golf. What we never imagined is that we'd actually be changing people's lives.
The best customer experience is the one your customers never have to think about.
Hi, Sam. Sorry about your train cancellation. Let's get you booked on the next available train.
Because you communicate proactively, anticipating their needs before they become needs. Because you deliver 24/7, always-on service with AI Agents that offer personalized experiences. Because you can scale with agentic AI that completes tasks and resolves customer issues in the languages and channels customers prefer. Because you empower customer-facing teams with AI tools so they can stay engaged while providing higher-touch support. Because you harness AI to turn insights into action and enhance every part of the customer journey. Experiences so effortless and engaging, your customers keep coming back. That's the power of Webex and AI-first customer experience. Webex Calling Customer Assist enables your local and regional employees to deliver seamless customer assistance with AI-powered features in the collaboration app they already use. Click to call allows your customers to get in touch with one click. No login required. Once they've reached out, any team member can answer and help.
With AI Assistant, they'll have efficient tools at their fingertips, like real-time call summaries that can be shared when transferring a call. With the call sentiment for supervisors feature, supervisors can assist team members whenever they need support. Enable your local and regional employees to deliver seamless customer assistance empowered with AI in the collaboration app they already use. Webex Calling Customer Assist.
Cisco and SAP have agreed on ambitious milestones for the migration of our more than 250 offices globally. Right now, we are planning on having upwards of 350,000 devices on the network, including IoT devices. We have more than 100,000 employees. They need best-in-class IT infrastructure to work wherever, whenever, and however they choose with the devices they prefer. With Cisco, we've made our network simpler and more secure. Cisco SDA also enabled us to improve our zero-trust access policy, ensuring only trusted devices can access corporate resources. With Cisco Spaces, we could leverage this foundation and establish a global platform for smart offices. Our first experience confirms our hopes and expectations that we can manage everything in one place together with a professional partner like Cisco.
Tribeca has always been curious about technology. I feel that so much of Webex and Cisco specifically have done for us is help us lower the barrier of entry so that all artists can participate and be involved in the festival.
We're using Webex events in order to make the experience of De Niro Con much more interesting, immersive, and connected for our audience. A frictionless experience was key. We don't want our fans and customers worrying about how to find things or where to find things. All we want them to worry about is how do they have the best possible experience. We want to continue to tap into the broader suite of capabilities that Webex has, including a lot of the AI capabilities that they continue to build out.
It's important to wor k with.
Welcome back. We are here live at the Cisco Live TV studio right here. Did I say live? Yes, live, right here in sunny San Diego at the San Diego Convention Center. I tell you, we just wrapped up a fantastic session with a deep dive with our leaders here at Cisco, Raj and Kevin. They talked about a new approach to digital resilience, get ahead of the issues, and minimize business impact. They talked about how our organization and how your organization can explore this thing of digital resilience, how you can strategically manage and prioritize issues. We all want to prioritize our issues to reduce any types of disruptions, any type of latency issues. We want to make sure that we are gearing up our businesses to handle those types of things. Kevin and Raj, they talked about some Splunk observability.
I tell you, say that three times. Innovations plus inspiring customer stories. We all love a good customer story. They showed how teams just like yours, just like yours, how they're using data integration, how they're using advanced analytics and AI to tackle issues right now, today, showing us how we can triage, how we can triage issues based upon the business need, based upon those needs, and not treating every issue as if it were urgent. You know, there is a message that we have on our email, "Your fire is not my fire alarm." They are teaching us. They were speaking and showing us how we can just do that. Now I want to go over to my fantastic, charismatic co-host, Lauren. I'm about to say Michelle. Lauren is over there on the showcase floor, and she's not alone.
Lauren, who do you have with you today?
Z, I am here with Richard from our industrial, leader of our industrial IoT team here at Cisco. Richard, please tell me, what do we have?
We've got a lot of stuff going on on the floor. Z, how are you doing? It's good to see you. We have a lot happening. Sorry, my voice is a bit hoarse. We've been talking nonstop. We'll work our way through this. Really, a couple of trends. I'm going to go through three trends, and then we're going to go look at some toys that we've set up in our booth.
Amazing.
What we're seeing in the industrial space, and when I say that, what I mean is manufacturing, the roadside that you're driving on, the electric power grid, how we manage that, ports, different types of the airport, different elements like that. What we're seeing is we're bringing those pieces onto the network. Because a lot of those customers, what they want is the ability to do kind of do three things. One is, how do I build a network policy so I can drive assurance? I can make sure I can find stuff on the floor.
Yep.
Two is, how do I secure that? How do I make sure that somebody doesn't hack into my plant and gets the intellectual property of my process that I have? Finally, what we've got is this developing area of machine learning and AI. This idea that we've got, we're moving back a lot of data. That's going to be our first demo that we're going to take a look at. We're moving a lot of data basically underneath the core network that liberates the automation network and makes that just operate more efficiently. Because when you drive latency into a production process, these tubs right here don't go around as quickly.
I'm excited. I cannot wait to see what exactly we're doing. Because I've been taking a look at this thing while walking through.
It is mesmerizing.
It literally.
Let's take a look at it.
I want to hear what's really happening.
Okay. This is just a small segment of a much longer production line. This was a food and beverage use case. What we have here is an ice cream vendor. We're working with Omron, the PLC maker, to really on this solution set. A couple of things here, what's happening. We have seven cameras that are scanning these tubs, and they're doing a couple of things. One is this whole traceability element. What they're looking for is, number one, do I have a mismatch between my top and the ice cream that's inside in the bottom? I need to be able to see that because I don't want that going out of the plant.
Exactly.
We also can hit this with a microbial sensor and look for different pathogens that might be on the ice cream. If it goes out, I can find it. I can do the recall on that.
Oh, from a safety perspective.
Exactly.
It's not just the efficiency.
Think about thousands of these things moving along. Right now, let's talk about the actual device and where Cisco plays in. We have seven cameras, and we're backhauling that data using our industrial switches that we're showing here. We just launched these, came on the market this morning, the IE3500 and the IE3500H together, and they're doing a lot of the backhaul of the data. Now, each of these cameras is running is basically a mini server. Okay? They're generating a lot of data. We need like 10 gig to move that data along the nervous system back into the actual containers that tell them what's going on here. That's a trend that we're seeing across the board.
This idea of visual inspection, we'll talk about it again on this guitar demo, but that's a key use case, first mover use case that we see in the industrial world. Second point.
Yes.
These are HTTP servers on these cameras. For the operator, this device looks like an IT device, right? It's got compute embedded in here. I need to be able to build the container that can really manage that and move the data where I want to move it. Either into the cloud or a data center or element, and that's what we've got here.
Wow.
An idea of how we're taking a process and then bringing the network underneath it and then moving that data to the analytical piece.
You need the network.
Absolutely.
Essentially, the network is the foundation of what we're seeing.
If you didn't, this would be a completely isolated environment inside the plant floor. I can't, it's hard to secure. It's hard when something goes wrong. I don't know where that's happening. There are a number of things like that that we're working on.
Wow.
Okay.
Wow.
Let's dig. Are you ready to step it up?
Let's step it up.
Okay. Let me.
All right. This is good. What's next? Let's look at the guitars. I think this is the favorite demo here. Everyone's been looking at these guitars.
Yeah. With this on the floor, it has been insane, this entire deal. All right. What we have is a modern production line that we've built with these guitars, except instead of making the guitars, we're tuning them. All right? We got a number of things that Cisco's embedded into this. Long story short, these guitars move around this track, and we scan them for defects. The same way that I was doing that earlier on that ice cream tubs.
Yep.
That's what we're doing with our guitars. Think about if you have a car manufacturer and they get molten aluminum on the chassis, and you got to polish that out. You need to know that. That way, I can pull the car out of the production line and give it special attention. Same thing that we're doing with our guitars that have some blemishes on them, right? First thing we're doing.
Which?
Second.
Because we can see that scratch even, right?
Yeah. Exactly.
Yeah.
Scratch that we don't want to sell that. Not only that, but we want to figure out where's that scratch coming from? Do we have a machine that's not lined up, and it's causing that defect on them?
Great point.
All right. This is our control layer. Key thing here, right? Normally, you would see human-machine interfaces and PLCs in this control layer. They're not in there. What we've done, and this is a trend that we're seeing in the industrial environment, is we've virtualized all those applications. We're running that in this data center. We have a UCS on this black.
That's running a virtual PLC?
Yeah. It's right there. Yeah. We run that, and then we can run that as a container connected into the I/O devices, and that is powering all the systems on the machine.
Amazing.
Much more. Now if I need to upgrade the software, I'm just doing a single upgrade on that system. I can either do it.
Less touch points, right?
Exactly. I can either do it on the UCS, or if I need to do a software upgrade on something here like these cameras, I can come right through and do that. We use secure equipment access. We build basically a cloud broker and come through on that. That is our kind of key.
CL3500s again, right?
Yeah. Exactly. Exactly. All right. ITOT convergence. Key theme for us. Who do we have here at Cisco Live? Most of these people run IT networks for these industrial customers that we work with. What we have is a dashboard that we've done with Splunk. A couple of things here. Number one, we use Cybervision to map everything that's in this production line. Now, think of this if this was a 100,000 sq ft plant. You have thousands of IP addresses and devices and sensors. That all gets mapped. We discover that, and we're now bringing AI to offer a segmentation solution set there. We do that. We can automatically do that because that takes a lot of time to go through all that data. Now an AI agent is doing that for the use case.
You can see we've got the production network and the infrastructure network. Great advantage with Splunk because what we're doing is we're taking all the production stats, all the key performance indicators, the KPIs, and we're matching it with how the infrastructure is performing. Perfect use case for Splunk where I've got different types of data sets, and I really want to optimize both the production environment as well as the IT network that's fueling that, the security piece that's fueling that as well.
I love it. That's one Cisco right there.
Yeah. Absolutely.
The fact that we are really not looking at these as separate technologies, but how do they work together to build this outcome that we see?
Right. Because our switch, so we just launched our 3100 switch for this machine robot.
Ooh.
We just cut in.
That's cool.
With these connectors right here. Yeah. Small form factor, but again, so it can work with a machine. It's not big. Then we can backhaul it back to the switches you saw there and then to the data center that's on this side. When you look at.
Okay.
The data center, right, it's hosting all the AI modeling, the virtualization, the IoT segmentation, Cybervision for OT, all of that's on this data center piece. You can see our switching here and here as well.
Oh, wow.
To connect all of that. This pace of automation is unprecedented because what we're seeing is, and especially with the disruption that we're with either the electrical grid or with manufacturing and where things are made, how do we spin up a manufacturing environment very, very quickly? I need to do that as a kind of a cookie-cutter set of policies that I can push out. If I've got a global operation, I want to build the plant the same every time.
I love it.
That makes sense.
It's almost like a shop in a box, and I'm just saying that, but it's all condensed and simplified. What I'm hearing is just.
It's like totally converged the IT and the OT network together, kind of work through that. All right. A lot of times, it's not just about our connectivity pieces. We got to talk wireless, right? Because I can't have wires all over the place. Let's go over here and take a look at this piece over here. We launched, and we'll stay out of our friend's way here. We have I don't know about you. I love cookies and cream ice cream. I get a little bit of chocolate. I have a little bit of vanilla. That's what we're doing right here with on Wi-Fi. Okay?
Oh, okay.
We have a proprietary.
That's a great analogy, by the way.
Yeah. Wait, wait, watch. We're going to blend this together. Okay. We have been making industrial APs, and they use a very specific channel to do low latency, high fiber-like connectivity. We use that to connect trains, like the track side of the train or different types of environments where I can't break that connection, right? We have regular Wi-Fi, right? Everybody knows Wi-Fi. I've got it on my phone. I'm connected on that. When you start blending that or you start thinking about Wi-Fi, you run into a density problem. Because you all of a sudden have a lot of things going on in the plant.
Yep.
Let's use that example I did for the machine or the car that got the molten aluminum on it. That's on typically that might be on a robot. So I need to be able to communicate that and move that robot around. I can't use the Wi-Fi network for that because I got people looking at YouTube videos in the plant.
You need something dedicated.
Right. Now what I don't want to do is do different types of APs. Here comes our cookies and cream ice cream. Because what we've done is we've taken ultra backhaul through Catalyst Center, and now we're running that on our Wi-Fi and our WinBo APs. You can do mix and match all with the click of a button in Catalyst Center. It is much more expansive. I can get all of my same APs.
Yep.
We're not putting anything new in the plant. Same APs. This AGV, it needs to be on ultra backhaul. Check the box in Catalyst Center. I've got control systems that are on an iPad, and I need to be able to do that as well. That may be on the Wi-Fi network. Those keys are kind of what we're working on there.
I am extremely impressed and excited, and I can't wait for customers to see this.
Oh, yeah.
I know they're already standing in line.
Yeah. We're looking at this. Very up here all day, all night with all of our stuff. So it's really exciting.
It's such a pleasure to be here with you.
Thanks.
Thank you so much for your time and for explaining all of this to us. I'm going to send it back over to the studio to Steve Moultrie.
Bye, guys.
Lauren, Richard, I tell you, Richard, you hypnotized me with the assembly line, and then you woke me up with the cookies and cream because you're speaking my love language right there, sir. We are here in the studio. I need you to hold on to your seats because I'm going to throw it over to Steve because he's in the studio with some amazing guests. Steve, who do you have with you today?
Thank you so much, Z. I do appreciate it. It is time right now that we get to talk Intel, which is always one of my favorite things to do here at the show. Intel, one of our longest-served, most deeply valued, and important partners here at Cisco. I've got Dorin Vanderjack here in the studio with me. Dorin.
Good to see you again.
VP, General Manager of US OEM, and Strategic Account Salesman. You've got a long title, Dorin.
Yes.
Glad to have you here.
Yes. Thank you. It's great to be here. Thanks for having me again.
Oh, absolutely. Such a pleasure. I've got Mike Bundy over here with us. Mike is Senior Director, Global Cloud and AI Platform Sales with us here at Cisco Compute. Michael, I don't know that we could have put together a better duo between you and Dorin here at this show.
It's awesome. It's awesome.
It's a power play, right? Again, it's all about the partnership. All right. So, gentlemen, here's what I want to do. I want to kind of start with data center modernization because it's such a big topic here at the show. We heard about it this morning in the keynote from Jeetu, from Chuck, how we're powering the AI era. And we know the challenge, right? AI becomes more mainstream by the moment, not by the day. We have this pressure to modernize the infrastructure, right? It's intensifying at such an alarming rate. That drives compute costs. That makes sustainability targets harder to reach. It makes agility harder to achieve, right?
Yep.
I want to talk to both of you how Intel and Cisco team up to tackle this industry need. Dorin, let me go for you, and then, Mike, I want to come to you as well on the Cisco side.
Sure. No, hey, first, thanks for having me again. It's great to be here. When I think about how much the industry has pivoted to take advantage of AI and its infrastructure and the things that you really are concerned now more about, which you never would have thought of, constrained environments, do I have enough water? Do I have enough power? For crying out loud, we've got nuclear mini mills now that we're putting in the middle of deserts to power infrastructure. I think about the introduction of the new Xeon 6 and what we are trying to accomplish there and what we've been working together with Cisco on for many, many years. That is, how do I pack more punch into an existing platform? I think about things like workload consolidation, i.e., data center modernization. How long have we been talking about this?
We just have a new name for it. That's packing more punch into the same footprint. From a power perspective, we actually get better, more workload, and you really want that capability in one device because you don't have a larger footprint. You can't afford to put an adjacent GPU in. In that footprint, you really are running AI workloads. We put in AMX and MRDIMMs to provide the capability to do inferencing, to do workload—sorry, do the inferencing and small language model operations within the same CPU. A lot of capability. The other piece I think about is the optimizations because it comes back to sustainability. We've worked with common application platforms like TensorFlow and PyTorch to optimize those. We've looked at what does a customer really need. If I want to modernize my data center, I want to do that at my own pace.
This gives you the capability to do it at your own pace. Everything we've done with our partner Cisco has been, let's deliver a future-proof platform on these open standards that allow you to modernize your data center at a superior TCO.
Love that.
In an open fashion.
Yeah. So important, right? It's that whole idea of how do we minimize the footprint and maximize the investment, right, Michael?
Absolutely.
From your perspective as well, build on what Dorin just talked about.
Modernizing the data center, we've been doing that together for 15 years, actually.
Yeah. I love that. New name.
Yeah. We've revolutionized it together back in 2010. We look now with what you guys have done with the Xeon 6P processors. We've had thousands of customers that have made the upgrades in the last 12 months with us. It's done wonders from a sustainability point of view, from a cost savings point of view. We can get, like you said, more processing power in a much smaller footprint than we've ever been able to do before. The innovation that we do inside of our server has allowed us to pivot. How do we want to use the real estate? Is it a CPU-centric solution, or is it something that's hybrid with GPU? We're able to create that flexibility within the platform. A lot of it's because of what you've allowed us to do with your superior technology.
It gives us that affordability to be able to make those opportunities for customers.
That cross-empowerment, right? Right?
Absolutely.
Love it. All right. Let's turn a little bit. I want to look at security, right? Because it's got to be embedded in everything that we do, and it has been from the beginning. Whatever we build together, trust here in the age of AI is so important. Hybrid cloud security is, I feel like it's evolving as fast as the workloads that it has to support. They're not separate. What do we do? How do we address the challenges of secure data, but make sure that secure collaborative AI development doesn't get impeded? They have to work together, right?
Yes.
Security has to be native, doesn't it, Dorin?
100%. 100%. I think when I think about security, it's one of those additional things that come on board the hardware in this case that we've partnered very closely with Cisco to deploy. I'll start with the hardware, then I'll talk about the implications from the software and the end customer usage model side. We've talked about the explosion in AI, but with that comes existential threat for security. Why? Because the most important data that I need involves sharing some element of my data. Onboard the processor—yes, back to the processor. Sorry, that's what we do. We have SGX and TDX extensions. Those are extensions that allow us to create a virtual vault with data in motion. That means your most important data is essentially safeguarded. Social security numbers for very, very sensitive customer or HR data.
You've got national secrets, those kinds of things vaulted away. Very importantly, I still need to be able to do something with that data. The do something is, hey, search all of these social security numbers and find the ones that don't work out. Those aren't good, right? Or that person doesn't exist in this other system. With confidential compute, which is what we've implemented together with Cisco, this allows us to do things with data in motion while not compromising that vaulted data. It is a very important distinction, and it's something that we've put together to deliver at the edge with value that already exists in the processor.
Vital, right, Mike?
Yes.
Because, again, it comes down to the human. It comes down to the individual getting to the data, accessing what's taking place on the network. If it's not protected, it can't be connected. I've heard that somewhere before, right, Mike?
Exactly. Exactly. I think leveraging what you're talking about at the server level, it fundamentally starts with the network with us, right? I mean, as Chuck said this morning, this used to be called networkers in 1997 or 1998. Integrated security at the network layer is something we're keenly focused on. We have developed software for things like modeling. When you're building your language models, how do you ensure the integrity of what's being built is actually functioning the way it was designed to function with our AI defense platforms? It is a great partnership. We're secure inside the server. We're secure outside and getting to the data. Excellent point.
I sort of feel like there might not be any other type of partnership that could do it quite in the way that Intel and Cisco could do it together. Am I kind of grasping at straws there?
No, you're not. I think the biggest thing that we've done is over multiple generations of delivering this, we've kind of figured out the right lane for Cisco to leverage our technology and Cisco to actually build upon it in their own platform and then deliver to the end customers exactly what they need, more customized. Because let's be clear, we are an ingredient in this play. We want to enable the best technology, and that's what Cisco delivers.
Oh, I love that. I really do love that.
It is super powerful. I just looked at a customer deployment recently, and what we did together, you talk about sustainability going back to that, something like 45% savings from rack space and power based on what we were able to build together.
Isn't that cool? What a great story to be able to share out there.
Yeah.
Let's kind of push this out a little bit. Let's go all the way to the edge on this, right? If we look at the future of real-time AI, what we're talking about here, right? We've talked about ways that we can modernize the data center, the ways that we secure the infrastructure, but the edge is one of the fastest-growing areas of innovation that we're looking at, right? AI is pushing closer and closer and closer to where the data is actually being created.
That's right.
How do we see edge computing continuing to evolve in real-world deployments, and what does it mean for our customers, for Intel customers, for Cisco customers, everybody here at the show?
I can start it off, Mike? Okay. Okay. Very proud of the fact that we have been collaborating—Cisco and Intel have been collaborating—on edge since a little platform, not very well known, called Moon Island back in 2011.
All right. That was one. You even lost me on that one, Dorin.
That was the first time that we decided to drop a Xeon in an edge device. At that time, it was more of an access point. Everyone was up in arms. Why? Because they were very concerned that the compute capability would shift, oh, Lord forbid, out of the data center. Guess what's happened in the last? It shifted. Yes, to take advantage of all those myriad opportunities that exist out on the edge where you have to be able to perform an AI/ML operation, things like invisible fence, which is used in the military, things like predictive maintenance, which is used almost everywhere—think train cars, think water towers, think critical infrastructure—all the reliability pieces. You need that capability sitting on the edge.
Cisco has been absolutely adroit in addressing those specific opportunities point by point with customers, whether it's been in the rail, whether it's been industrial, whether it's been healthcare. I think what we're doing is exposing back to the infrastructure that we provide OpenVINO, one API, to allow these kinds of operations on the edge. This is so cool.
It is just something where we have just had an absolutely killer partnership. I'm super proud of it, and I think that's going to continue forever.
This is something to be proud of, isn't it, Mike?
It is. I think the co-innovation we're doing, we're going to release a new platform soon that will have integrated storage, integrated compute, integrated security, integrated wireless, very, very flexible in terms of the use cases that customers could have. We've got some retail customers that have tens of thousands of locations that are looking at this as a game-changing platform. It really goes back to the innovation that we've done all along, like you said, for the last 14 years that allowed us to leverage what we have and then add the additional components into it. For us, it's going to be a game changer because you see in the industry solutions that are focused on secure access or secure edge. It's got some networking. It's got some security, but not all the other components that a customer could drive so much cost out.
Just the pure savings from an OpEx point of view, when they have a problem, it allows them not to have to send a truck. It is going to be game-changing when it hits the market.
We have got some examples down on the showroom floor together.
Thank you.
I was going to ask you. There are some in the Cisco booth. There are some in the Intel booth. You got customers, our collaborative customers like Wait Time down there. There are some just amazing demos. You have to kind of see them to go, "I didn't know you could do that." I mean, think about self-checkout, kiosks, monitoring for things where you don't have an RFID or a code. It's these kinds of usage models that will continue to evolve. You get to walk out of a Sam's Club without somebody rifling through your bag. It's kind of a good value prop. You'll see these kinds of demos. That is why you come to Cisco Live, is to be able to walk around, to be able to get to play with them. It's sort of an advanced.
Absolutely.
Right? Gentlemen, like I said, I really think that the nature of this partnership, it's proven itself out over time, but it's when we get an opportunity to sit down together like this and really deep dive. I mean, I can't believe that you brought us back to—what is it called again? 2011?
Moon Island, 2011.
I'm going to look that up. Moon Island 14 years ago. Guys, thank you so much, Mike. Really appreciate everything that you're doing. Dorin, really grateful to you for the partnership.
Absolutely.
Great conversation. Please come back and visit us again here in the studio, all right?
Thank you, sir.
Thank you, gents. Zee, let's send it back over to you.
Wow, Steve, I tell you, if you are not here, you are missing it. This must be how Chuck feels all the time. I feel like a proud mama. I like tearing up over here, over here with Dorin and Michael. I tell you, this is amazing. Coming up next, the branch is evolving. You want to lean in on this next session. Three things you need to expect. Let me tell you what you need to be listening for. Get back to my notes because I do not want to miss this. Integrated solutions, you need to be listening for that. Simplified management, advanced capabilities. That is what you are going to hear in this next session. Guys, go get a cup of water. Go get some coffee. Go get some tea. You need to lean in on this next session.
I tell you, I can't speak anymore, Steve, about this last conversation. I'm over here, and I'm feeling all sentimental, and I'm like, "I work here. This is my company." I'm feeling like a proud parent. I mean, I want to go over here and just hug Michael and Dorin, and I know that's so inappropriate. I tell you, we are getting ready for this next session. We have speakers, Chris, Ralf, and Lee. They are going to crack open this nut about branch, unified branch approach. That's a design to simplify management and enhance performance. Is that what you need as a customer? You need to be listening in on this next session. I'm going to pause right here and just take it all in, the energy that's in this room.
Cisco Live, you missed it this year, but I tell you, like Chuck says, we say this every year, this is the best year. I tell you, next year, it's going to be even better. It's going to be so much better. Like I say, coming up next, the branch, you want to watch, you want to listen, you want to take notes, you're going to hear something, and you can go ahead and say you heard it and you thought of it yourself, but you really didn't. Lean in. This next session is coming up right now. We'll see you back here in just a few.
Thank you. How's everybody doing? How's the day going? I fell off the stairs earlier. I got on stage without falling on the stairs, so already, this is a win. It's only going to get better from here. That's amazing. Look, thank you, everybody, for taking the time. I am here to talk about the branch, the branch operations. My role at Cisco is focused on the WAN edge, the secure WAN devices that we launch. They're a very important control point for us as we think about branches, right? Branches are super critical to our customers' operations, the ability to meet your customers where they are, to do a job of delighting them, to be able to interact with them, to be able to transact with them. If that doesn't work, that's going to be the number one driver of customer dissatisfaction.
With so many choices for our customers today, you run the risk that the loyalty and the bad things that come from having a bad customer experience, a bad employee experience in those branch locations. We see some trends when we talk to—and it's very universal. It doesn't matter whether we're talking branches in India, branches in the U.S., branches in Europe. We hear the same things from customers. It's getting more complex, right? The things are not getting simpler over time. The amount of applications and where those applications live has proliferated into so many different places that it's not as simple as it's been in the past. That drives a lot of the complexity and drives a lot of the challenges.
I'm going to get into some of the techniques we at Cisco are bringing to bear in order to try and simplify that, to try and make that easier. Security is front of mind. There was a view that security and networking were somewhat separate. We are innovating and making sure that we treat those as one thing because even if somebody is only responsible for the network, I guarantee you there's somebody else in the organization responsible for security that's going to have a great deal of involvement and have strong opinions on what actually happens from that networking device perspective. Being able to make sure we deliver secure experiences is very important. The other universal trend is the skill shortage, right? We hear this everywhere that there's just not enough skilled individuals.
Our job as Cisco is to help make things simpler, to make you more efficient so you can scale, so you can focus on the things that matter, not on the routine tasks associated with running networking. Finally, AI pressure. That's coming at a number of different angles, right? There is the application of AI in your own networks and the things that you're doing to service your customers and service your employees. There is the rising threat of AI from a security perspective, right? The AI renaissance is in front of us in terms of the ability for bad actors to use AI to come up with new attack vectors on our networking infrastructure. It's very important that we deal with that.
What this all leads to is that we need to start thinking about how we architect our networks, how we architect the branch differently. That's what I'm here to talk to you about today, is how we're doing that. I'll cover that in three areas. This was the same areas in terms of the pillars that we talked about in the keynote this morning: operational simplicity powered by AI, security and the networking being fused together, being one and the same, and how we build devices that are scalable, that are ready for the AI era. Starting off with the simplicity point, within the networking portfolio, we've spent a lot of time thinking about how do we make things simpler. We don't make things simpler by building new things that don't integrate with each other.
Starting with the hardware, introducing a lot of common hardware. I'll get into more specifics of what that means for the WAN portfolio later, a consistent way that we license and that we have entitlement for using these products, and then finally, simplifying the management and converging that as well. It means that you get the best of every world, right? You want to use CLI mixed in with cloud. You want to do some on-prem. The ability to operate those in a consistent manner is a big part of what we've been working on and what we're very excited to launch here at Cisco Live and be able to talk more about as we go through this event. We are also thinking about not just how do we save, but how do we make it real, right? The platform is really where it all starts.
The ability for any of these devices, whether they were traditionally what we would have said was a Catalyst device, what would have been a cloud-managed Meraki device, a consistent place to manage those from a control plane perspective, which then becomes the point at which we can expose APIs for interacting with it. It then becomes the place where we expose analytics and intelligence in terms of being able to interact and troubleshoot those networks. It's also where assurance comes in. Our ability to look at our networking estate, owned/unowned infrastructure, and be able to help pinpoint problems. A lot of the times, the number one thing our customers ask for is just, "When there's a problem, just help me find it. Actually, just practically tell me there's a problem.
Tell me what I've got to do to fix it." And that's the things we've been able to build. We have more devices on the planet from a networking perspective than anybody else. We have the biggest data lake. We have decades of experience from CCIE. Being able to take that knowledge and apply it at this platform level drives great things for us. I'm going to get into branch's code in a moment and what we're innovating and what we're announcing there. Unified management, I've touched on that already. Consistent management of Catalyst and Meraki devices in the cloud. And then the end-to-end assurance.
Baking ThousandEyes as an assurance experience into all our devices, those devices being able to act as endpoints and being able to troubleshoot and be able to be the points at which we run synthetic tests to be able to tell what's going on in the network to build a very rich picture of quality of experience and what's going on there. Getting into some detail around branch's code. Show of hands, who's heard of Terraform? That's more than the last. Percentage-wise, that's more than the last presentation. That's about half. That's great. That's great to hear. Terraform is an open-source set of tools that are able to be used for automating how we do things with pieces of infrastructure.
Think of the ability to take something like a network configuration and treat it the way and save it in the way we do software, right? The code that we do, that's the infrastructure as code. A lot of our effort has been around how do we take that as an approach. It's an open-source approach. It's freely available. You could go use it today. What could we at Cisco do to make that better? There's a few things we've done within that. One is that we've built validated designs, Cisco-validated designs around how you not only the equipment you use and how you deploy it in the branch, but also how do you interact with the actual platform itself.
We've also worked with our CX organization to make sure that we can help you on this journey if you decide that this is something you want to do. I'll get into the specifics of what that looks like with a demo here in a moment. We have a whole service offering around how we can actually train you on how to use this and take all that intelligence. I mentioned that decades of experience, right? We really, really shouldn't let things get misconfigured because we know the best. We know things that would actually cause network-breaking changes. The ability to actually look at the change before it gets deployed and verify that it's not going to actually cause damage and be able to control that as well. I'm going to talk about a scenario here. Imagine I've got three active sites.
I'm operating the network. My boss comes and says, "We need to deploy five new sites." Take your choice, retail sites, healthcare clinics, doesn't really matter. What would happen in the traditional world would be I would have to go into a user interface and I'd have to make lots of clicks. Maybe I don't have a management interface and maybe I'm manually configuring those devices on-site. It doesn't really scale because it might be five sites today, it might be 50 sites tomorrow, it might be 500 sites the week after that, right? Being able to do things at scale in a consistent way that is able to reduce the risk of mistakes, be able to reduce the errors, be able to reduce the security vulnerabilities you introduce when you're doing things manually.
I'm going to show what an infrastructure as code and what a deploy pipeline looks like here. This is in GitLab. This is our branch configuration stored as code in GitLab. Again, the same way that you would manage software. I'm able to use the same tools I would use to manage software here as well. I want to jump in. I'm going to make some edits to this. I'll go into the IDE here and actually I'm just going to copy the configuration, be able to say, "Give me five more sites that have that same configuration." That's where the consistency comes from. Now, because this is, again, part of a continual integration, continual deployment pipeline, all the things that are associated with that happen inherently. I'm going to push that back into Git.
I can see here it's actually running as a task as part of the pipeline. We can do some very clever things here as we do this. We can make sure that we don't commit that change to the actual deployment, to the deployed network before it's even pushed out. Let's confirm that it works. Let's test it. We could set up pipeline flows to say, "Go deploy that in a canary site and validate and verify it first." Let's get in front of that and make sure prior to the change that we just make sure that everything's okay. Also, when we see here where we've gotten to the end of that, after it's deployed, let's go check and make sure it didn't cause any detrimental impact.
We can go look at ThousandEyes and intelligence from other places to make sure that the change wasn't detrimental. What we can see here is after that pipeline is finished, I can now go into the dashboard and I can see I've now got five sites configured. My next five sites are logically deployed. What that then allows me to do is that if I've got those five sites and maybe they're on the four corners of the planet, I don't know. I can send the equipment that I'm going to deploy there and it becomes a plug-and-play experience. The type of person that you have to go and deploy, you don't need a CCNA, CCIE to go on sites with this configuration. You just need somebody who understands low-voltage cabling and is able to get these devices cabled up and running in the environment.
It massively simplifies it, massively makes sure that the risk of error is reduced, makes sure that you're not putting a box that day one is exposing your entire network because a tired network admin forgot to copy the right part of the configuration over. We drive a lot of best practices and a lot of the intelligence that we have into how we do this. Now, talking about the management plan, that is clearly very cloud-centric. There is one thing I want to make really clear. I hear this consistently from certainly some of our more experienced network admins. You can have my CLI when you pry it from my cold dead hand. We're not taking your CLI away. It's not going anywhere. We want to make sure that we have an optionality depending on where you are in this journey.
If you've got tooling and you've built systems and you're doing scripting and configurations that rely on the CLI, that's not going anywhere. That's powerful. It's massively powerful. This is one of the most established and the most widely deployed network operating systems in the world. We are not taking that away. We also see, whether it be regulatory reasons or whether there's other reasons that customers want to run on-prem, that's still incredibly valid, incredibly important to us. That's not going anywhere. Where we're looking at it is where does the majority of the customers lie. As more of the applications have moved to the cloud, we believe, and we certainly see this in the data in terms of deployment, that the simplicity and the ease of deploying from the cloud, the benefits of that are huge.
We want to be able to make that that's an option. Again, it's choose your own adventure. Not everyone will end up on the right of this chart, and that's okay. We're not taking any of the existing tooling away. It's also super important as we think about deploying in brownfields, right? This isn't throw your whole network away and start from scratch. We want to be able to make sure you can mix and match equipment. Maybe you've got some new sites. Maybe you've got some legacy sites. Recognizing that, and as I talked about the CVDs, the validated designs, they're going to be documented in that way as well. Now, quickly on the security, we want to make sure that as we progress, we've gone from MPLS circuits to SD-WAN. SD-WAN's evolving to SASE.
We see the next generation of that being over to Universal Zero Trust, right? Making sure that it's each step of that journey. Again, this is not mandatory that everyone's going to go to the right of this chart, that we've secured things in the right way, that we've got the right policies in place so that identity matters and the ability to recognize what type of applications that user is accessing. The network has an incredibly important role to play there in terms of it is the most important enforcement point within that policy. The other part of what we're baking in here is making sure that we have great experience with SASE, right? We would love that to be Cisco SASE. We would love this to be secure access.
We also recognize that there are third-party SSEs out there and other companies that customers may have already invested in. We will make sure we still enable a great experience for those as well. Finally, this was talked about a little bit this morning, but the idea around we think about personas. There's a networking persona, the security persona. Sometimes they're the same person, but often they're not. Making sure that we have an ability to address both those personas. A tighter integration between the security cloud control and what we do in the Cisco networking cloud to enable the cloud control to be the central point of definition of policy, but the networking component to be an enforcement point as well. Imagine a scenario where maybe I've got some Cisco Secure firewalls.
I'm using secure access, consistent policy deployed in secure cloud control that's able to be propagated down through the network control to the network devices. You're not having to configure that in two places. You can also define that policy natively in the network cloud. We do see, again, as we move more to cloud security, that being important to do in multiple places. Final thing I'm going to talk about here is just the new devices we're launching here at Cisco Live. We're really proud, very excited to launch the Cisco 8000 Series Secure Router. That's available for ordering now, shipping June, July, August, depending on the device that you order. It's a full refresh of the range. I'll get into that a little bit in a moment. Some real innovations we've brought to bear here.
We've got at Cisco we've invested a lot in building a secure networking processor for this device. That's really important because one of the main purposes of this device, one of its main jobs, is to accelerate cryptographic workloads. That could be IPSec tunnels. That could be SD-WAN. That could be things like looking at firewall services and having a processor that's powerful enough to be able to run the firewall services without too much of a sacrifice of speed from line rate. The secure networking processor is a very important part of how we do that. The throughput is very important. As AI workloads explode, people doing more in the branch, more reasoning at the edge. It's important that we have enough bandwidth that's consistent, low latency, reliable in order to be able to support that.
We're also seeing in certain industries, certainly financial services, certainly in federal applications, the concerns around quantum processing and what the availability of those looks like. There's some very bullish estimates that maybe we're five years away from a quantum chip. We may be more than that. Either way, you need to make investments now that are protected so that when these quantum computers that can do massive levels of parallel processing launch and they have the ability to decrypt all the things we're currently doing from a protection mechanism perspective, that you've got a hardware platform that's ready to support that. That's in there as well. Finally, just making sure I mentioned the firewall already. I've mentioned all the things that go on from the perspective of the cloud control, how we manage policy on these devices. These devices have to be security-first in posture.
That's exactly what we've built and what we've launched here. We also launched the Cisco Firewall 200 Series. You can go see this in the branch vignette over that direction. I mean, you get a moment. This is a purpose-built box built for certain applications where you maybe want on-box decrypt at the edge. Part of that wider ecosystem can be managed by that security cloud control. You can have that consistent policy between this and other networking devices you have in your network. Looking at the full range of products we've got here, we've got the full refresh, the 13 new SKUs we launched from the 8100 through to the 8500, the data sheets and the things that are available now.
There is a call to action at the end of this for you to be able to go check those out. Like I said, please drop by the booth. We have them all featured on the hardware wall. One thing I want to assure anyone here, though, is that the existing portfolio, this is an expansion of the portfolio. We have no announcement of end of sale of any of the existing equipment. If you are using MX today from the Meraki perspective, continue to use MX. If you are using Catalyst 8000 series, continue to use those. There is an amalgamation that will happen over time as we refresh those devices. The operating modes you use today, the software that you use today on these devices, the SD-WAN you run, those are not changing as part of this.
This is just a refresh of the portfolio to bring some of those generational improvements from a speed and throughput perspective and from a performance perspective there. I'm really excited now. Enough from me. I'd love to hear from one of our amazing customers. I love Cisco Live. One of my favorite things is hearing all the amazing things our customers do with our products. It blows me away every time. We have Nestlé, who we've been partnering with for a long time. Nestlé's mission is to unlock the power of food to improve the quality of life for everyone, right? That's amazing. I feel good knowing that my products are supporting that mission. We are very, very privileged to have Nestlé trust us to be the ones that are delivering that goal.
I'm very excited to welcome up to the stage Chris Rongarajan and Ralf Hubenthal to talk a little bit more about how they use our technology.
Hey, Lee.
Thanks.
I think we can see it, right?
Yes. Let's take a seat. Hi, everybody. I feel really, really excited to be talking to Nestlé. And I don't know if you guys know, but my goodness, the Nestlé brands, right? I looked at this and I thought, hey, if Nestlé was to stop functioning for some reason, please don't. Our grocery store is going to be empty, Ralf. This is true.
Pay for my salary, right?
The best of all, I see Maggie on here. I think you kept me fed throughout my childhood, so really excited. The only thing is that Lee and I had a battle. The Indians feel like the Maggie, you know, the Maggie Masala, yes? He said there's a chicken version. I'm like, I don't know. It's only one good version.
Look, we are the most global local company you can find, right? We cater to the individual flavor needs, cultures, and nutritional needs. Yes, in India, MAGGI Masala is massive. In Australia, we have chicken. We have beef. I can tell you my kids who grew up on Australian Milo, they would not drink Malaysian Milo. The difference is really one is granular, one is powder. They also would only eat the Australian version of Maggi. It really depends on your taste profile.
It makes sense that Ralf is from Switzerland. That was the most diplomatic answer. The truth of the matter is 2,000+ brands, 300 + factories, 275,000 employees, 1,700 production sites. Are you sleeping? Does sleep happen somewhere?
Yeah, actually very well. Thank you.
Thanks to Cisco. Talk to us a little bit about, I mean, this feels like giant complexity when you're going from carpeted floors to manufacturing floors to retail floors. How do you manage all that complexity?
By simplifying, right? Keep it simple. That's where it starts with. In the keynote we heard from Novo Nordisk, where they spoke about standardization, right? We drive a lot of standardization. We try to focus to differentiate where it matters on the consumer interaction. We do not need to differentiate, for example, on networking, right? Networking is the same. Let's be honest about it. We have maybe different beliefs. This is more like we are techies, right? We like our ways of doing things. In general, how I do networking, you can standardize much more than you think.
That's incredible. I mean, at Cisco, and I'm sure all of us really do care about the network. I'm just curious, can you talk a little bit about the partnership you've had with Cisco? How have you modernized your factories? How have you modernized your branch sites?
Yeah, so Nestlé is going through an incredible network transformation journey right now. For 20 years, we hadn't done much. Over the last two years, we basically took every single part. One big piece of it was our wide area network. We went really to a Lego system where we have five regional contracts for the last mile delivery, right? That's where we have all tier one providers. They really provide the cable, right? We partnered together with Google on the creation of Google WAN, where Cisco is also an important partner to Google, which we announced three weeks ago. We have a service integrator called Singtel, which is really bringing it all together. We have across all Cisco providing us the technology to tie it all together. We're rolling that now out.
We were finished with SD-WAN before we took it all apart. That was one of the reasons we actually can do it. Finishing our SD-WAN rollout, which was already based on Cisco, was an important milestone, which really allowed us to go down that journey. We also learned a lot. We challenged all your resiliency numbers you gave us. You cannot have more routes. We doubled it. It still worked. When we tripled it, we had a few outages. Do not do it. Double is OK. Triple is a really bad idea. Now we're going, for example, for multi-region. Something you came up new with. That is really where we drive the transformation together, where we challenged you, where we use our scale, and we work together in a partnership to drive and enhance performance for Nestlé.
Obviously all your other customers benefit from it as well.
We did pay him to say this. No.
No.
You know, one of the stories you shared with me is you had an undersea cable cut that really kind of reshaped. Can you share a little bit of that story?
Yeah. Look, around Africa, there are two routes undersea cables go, right? One was cut by some pirates. The other one was cut by some seismology, I guess, an undersea earthquake. There was only one cable working, which funny enough was a Google backbone, which we were negotiating at some point of time with them. Thanks together to the simplicity, we were able to reroute our traffic really fast through the Google backbone. We were up in pretty much less than a day.
I mean, and other people were used, I mean, I'm quoting. It took others about two weeks. There were a lot of people who were.
It depends. It depends on the country, right? Because West Africa has so many countries that were impacted. Yeah.
Can you talk a little bit about ThousandEyes? I know a lot of this visibility came from ThousandEyes' platform. Can you talk a little bit about that?
Yeah. Look, we are constantly driving operational excellence. Outages cost us a lot of money. We have factories when I'm down for more than eight minutes, I roughly have to throw away $2.5 million worth of raw material, right? It's not only about the money. It's also about the food waste, which we really try to avoid. Visibility and availability and resiliency is super important for us. We have something we call Monitoring What Matters, which really allows us to see from the business outcome all the way down to the last router and switch to see what's going on across our environment. If I see there's a drop in the factory production, I can straightaway look, OK, this factory is connected this way. I can go all the way down to the endpoint. ThousandEyes plays a major role on it.
We have ThousandEyes agents deployed in all of our most critical sites, feeding into this dashboard via OpenTalons. Yeah, we have a presentation layer, which we are using.
You know, it's interesting you talk about this monitoring what matters. A lot of you might have heard a little something today about AI Canvas, which is pretty exciting. What do you think?
Yeah. I love it. Hey, lift it up to the business process level, right? It's pretty much nice on the network layer, but it serves the most important part of my infrastructure, right? The one which keeps me awake most, but also have to worry about SAP, Azure, and all the other stuff we are servicing.
I am texting DJ Sampath. I'll send you the email.
Yes.
We'll get that going.
He gets our requirement list. It's long.
I bet your requirement list for Nestlé is pretty, you know, speaking of requirement list, I know Lee talked a lot about branch's code. When you think about infrastructure for the modern branch, how do you think about that? What's next?
Yeah. Look, so we have different kinds of branches, right? It comes with our complexity. I have sales offices. If they are down, you know, go on the route and sell our product, right? Or use VPN, right? That's good enough. Then I have factories, right? I have to cater for all these different demands. The simplicity is therefore the most important part. How can I manage? We have a categorization in Nestlé where we look by site types. We have manufacturing. We have distribution centers. We have retail. Throw in espresso boutiques. We have head offices. We have criticality levels. To each of them, we have a reference architecture. This reference architecture, I really want to manage as code, right? One of the reasons, for example, we choose Singtel together with Cisco is that, yes, the cable is important.
The last mile is important. Really, Singtel together with Cisco sees infrastructure or network as code. That really is a game changer for us. It helped us with automation. It helped us with efficiencies. Every deployment we do to our new network model afterwards, every single site, we're reviewing what went good, what didn't go good. We continue to automate to improve the Terraforms and all the stuff we are using during the deployment.
OK. Here's something I want to share with all of you because I was thinking about Nestlé and how, and you have some state, I mean, you're one of the few companies who's like straddling a 150-year-old factory and state-of-the-art factory. As you think about managing both, and as you think about all of this, what are what lessons you've learned? What are things that you can share with all of us that are like, these are the few things you should think about the future of connectivity?
It's not about the age of the factory, right? It brings complexity. If you have a heritage-listed factory, try to put in a satellite dish or a 5G antenna on top of it. You go through a lot of challenges along that way. The key thing is worry about your resiliency. Be smart about your resiliency. That's really one of the key things we have learned. Make sure you're resilient. You can operate independent and worry about your availability. Be creative, right? Be creative, but be also simple.
I do find, I'm not sure if I'm allowed to mention this. I'm prefacing when I mention it. I find it really interesting that in your latest Purina factories in the United States, they have robotic dogs for a dog. Thank you. I was like, oh, that's very meta. The robotic dogs in the dog factory. Somebody made the comment, at least they don't eat the dog food. Can you share a little bit about how the new age of robotics, IoT, agentic AI is playing a role within Nestlé?
It plays a huge role. Doug, you can talk to him afterwards. He can tell you much more about what these dogs all are doing. Actually, we even get one into our test data center or test lab in our IT lab, where we simulate an average factory to ensure it works flawlessly. Look, agentic AI or AI, we are in that age. There are multiple things to consider. I really liked it this morning in the keynote. I also liked it during the Google keynote. More and more people wake up to it. You will not have any success with agentic AI or your shiny toys you can get with it will not work unless your infrastructure is ready for it. Number one, you have to do in parallel your infrastructure and then become your own best customer.
What we are doing with monitoring what matters, and that's why we are so interested in the AI Canvas, how can I use for my monitoring, for my deployment, more AI so that we are the prototypes for the business? To do that, our data in IT and in the business need to be up to scratch. If your data models are not right, if you do not have to find your data strategy properly, if your data are not clean, all your shiny AI toys will not fly. They will not work. It is like those old days in BI where you said garbage in, garbage out. Here you get super garbage out. Only then, if that all is in order, you can work with AI. In between, yes, you can take some technical debt for some use cases.
Focus on your user communities, the training of your user community, because it's a big change for them from clicking to using native language. Hey, I want to create a leave request. The really exciting part about agentic AI for me is, yes, in IT, it will be about creating efficiencies, do things faster, create more resiliency. In the business world, it will be a true game changer on improving how we do procurement, not only efficiencies, how we interact better with the customers and generate more sales out of it. That's really where the long-term beauty of AI is.
Gosh, that's really exciting. The last few seconds, this partnership with Cisco, if there's one thing that you've been most pleased about, what would it be?
There's a lot where I'm very pleased about. Like Aircan, our account manager, I'm really pleased with how we go about it. It's a partnership for more than 20 years. It's like a marriage, right? You have your ups and downs. Sometimes you shout at each other. Other times you're a lovey-dovey. For the last couple of years, we are in the lovey-dovey part again. Let's keep it there and have some co-innovation together.
Awesome. Thank you all. Before you go, I have two things to remind you of. We have amazing demos. Go see Branch's code. Check out all the different spaces. If you take a picture of the QR code, it actually takes you to Vikas Bhutani's amazing blog about everything we've spoken here. The most important part, Ralf is available for autographs and photographs right after this. Ralf, I really, really thank you.
You mixed me up with George Clooney. He does Nespresso, not IT.
On that, thank you, Ralf.
Thank you.
Hey, Edgar here, everybody. Welcome back to the Cisco TV broadcast studio. It is so great to have you all with us here on the real-time live stream coming to you from Cisco Live here in San Diego. I'm Steve Moultrie. We have just wrapped another fantastic center stage session. This time, we got the inside story on Cisco's new unified branch approach, all of the different ways that it integrates routing with embedded security and switching and Wi-Fi and ThousandEyes, puts the whole package together and then managed through a single platform dashboard. Lee Peterson, Dan Stewart did such a fantastic job showing us exactly how Cisco is addressing modern challenges. For example, cloud migration complexities and increased security threats like unmanaged devices, post-quantum cryptography. They talked about the different ways that we are driving WAN modernization for better bandwidth to meet the demands out across our multi-cloud environments.
So cool. So much great content. I want to give you a quick reminder to keep reaching out to us on social media. Please make sure that you use #CiscoLive or @CiscoLive. Right now, we are going to go out to center stage in the world of solutions, right in the middle of the showcase. We're going to be talking in just a moment with the men of the hour. Robb is going to have Daniel and Lee there with him. One of the other things they talked about was operational efficiency. So important. We had a great conversation about that a few minutes ago right here in the studio. Cisco helps organizations like yours to automate branch deployments so you get more efficient management, better scalability, maintain that high performance.
Right now, we are going to head out to Michelle, who is in the World of Solutions, with Jessie Reed. That is over at Cisco Customer Experience. Hey there, Michelle.
Steve, thanks so much for handing it down to us. I am here with Jessie Reed, the VP of Product for Cisco Customer Experience. I wanted to send it back to you right now with a demo of Cisco Unified Branch. He's going to walk us through this right now.
Hello, everyone. Great to be with you today. Unified Branch, let me get to the punchline right up front. What are we able to do? We're able to bring a complete solution together for our customers, for all your infrastructure at a branch site, and automate it end to end. What does that mean? Routing, switching, wireless security. All a customer has to do is express their intent, and we automate everything from there. Let me talk about why that matters. Our customers are dealing with great complexity. They have challenges with misconfigurations. Those misconfigurations result in security breaches, network outages. They're also dealing with having to perform tasks manually to a large degree today. What that means is that technology is advancing at such a rapid pace, and they can't keep up with the technology that they need their people to understand.
We can help them. They're asking us to help them in this many ways: simplified management, enhanced security, reliable connectivity, scalability, and automation end to end. That's exactly what this solution does. I mentioned routing, switching, and wireless security all coming together for a branch site. We do that first by using a platform-driven approach, just like Jeetu was talking about this morning. We do that with a dashboard interface. Next, we bring to the unified branch Cisco-validated designs. Think about all of the experience and wealth of knowledge that we can bring digitally to customers so that they can be successful. We use observability to make sure we're looking at application performance, end user experience, network performance as we look at this solution. We bring in branch's code. This is our automation toolkit that allows us to do that end-to-end automation.
Of course, no solution would be complete if we did not have support. Cisco TAC, one of our greatest assets, is integrated into this solution. If you want a managed service, we put a managed wrap around it with our partners. When I said earlier that all the customer has to do is express their intent, that is exactly what we mean. On the left-hand side are all of the inputs. We see critical apps. Customer tells us what they want to achieve, along with service connections, management services, and security. Look at all the elements on the right-hand side that are fully automated for them. They do not have to go in and program them. That happens automatically with the solution. Really, really simplified. I am excited to be able to show it to you right now.
Imagine that we're deploying additional sites inside an enterprise. We have three active today, and we're needing to deploy an additional five. We want to take this approach and simplify the approach. All I have to do is go into the dashboard and I say, create a network. When I go to the next screen, I type in the name of that network that I want to create. I identify the profile for that site, and then I select the create the network. Now all the configuration is being generated for the customer. All of those elements that I said that were automated on the right-hand side are being automated, and the configuration is being created for the customer. They no longer have to worry about segmentation and all of these other attributes because it's automated for them.
In here, you see the configuration as code. It's a repository. This is what's in production right now for the customer. We're taking this same approach, and we're saying, hey, we're going to go in, and we're going to pull in that configuration that was generated, and I'm going to add it to the environment. Just like code, where you check it out, I've checked out the code, and now I'm adding the configuration as code that I want to be introduced into the environment. The next step is what we want to do is add this to the pipeline. We're going to go through a methodical process to make sure that the customer does not make any mistakes.
We're saying, hey, I want to do a merge request of this configuration in production, and I give a little bit of a description, and the pipeline's executing. All of this is happening in the background. The customer doesn't have to go to any command line. It's all happening real time. We're going through, and now we're doing pre-change validation. This is where we're pulling in that validated design, making sure we're doing all of those semantic checks, those dependencies, making sure the customer is following best practices. When it completes that step, now it's the opportunity to execute it in production. We're pushing that configuration right now as we speak. This configuration is going into the production network in a fully automated way. Instead of me going to a command line and typing those things in, it's happening all right now as code.
Once that configuration is deployed, what you want to do is validate it. Is the environment working as expected? All of those rules associated with adjacencies and configuration parameters that should be functioning, we've now tested, and it's working. If I go back to the dashboard, look at this. I have the additional sites that I selected all configured within the environment. Extremely powerful, extremely simple. It's simple, accelerates your time to value, and brings our customers resilience.
Jesse, one quick question. Can you speak to the impact this has on customers?
Yeah. The impact is actually pretty tremendous. Our customers have always struggled with implementation success, and they're able to get upwards of 100% implementation success five times faster. As a matter of fact, one of our customers told us that they were able to achieve what used to take them 29 hours, they can now do in one minute.
That is incredible. That speaks to the power of this. One last question for you. This is big changes for the engineers. How are we helping at Cisco to help that transition?
Yes. This is a change for our customer engineers. They need to move to DevOps in order to be able to get the benefit of this technology. They want to understand the environment. They want to have control of the environment. What we've done is we've created assistants that help them understand the configuration, get information back about the state of that environment and how it's performing. We get to use that assistant to make changes to the configuration. As they make changes, it comes back and tells them, hey, use this configuration for the YAML environment. Finally, as we're executing the pipeline that I was showing you, there will be errors. Inevitably, there will be errors. The AI assistant troubleshoots those errors and corrects them for them so they can just seamlessly execute it in the pipeline.
All of that is done at a site level based on the customer intent.
Jesse, this was excellent. I learned a lot from you. Thank you. We're going to head back to Steve in the studio. Thank you guys so much.
Lauren, thank you so much. Excellent job there, Jesse. Really nice walkthrough on Unified Branch, on TAC, management services, automation around validation and deployment. Really appreciate that powerful user story over at Cisco Customer Experience. Right now, as promised, Robb is back out in center stage with the men of the hour, with Daniel and Lisa. Rob, let me go ahead and send it over to you, my friend.
Oh, I'm glad you mentioned Daniel's name because I was going to let it be a big reveal. Yes, I've got Lee here. Just came off stage, VP of Secure WAN, correct? Lee Peterson. We're going to have you follow up. You've got a special guest with you. I'm going to hand you the mic and let you kind of go through that, if you will. Sir, your mic.
Thank you, Rob. Yeah, very happy to be joined here by Dan Stewart from Ingram Micro. Dan, we'll just jump straight into the questions.
Definitely.
We've just finished a presentation about the Unified Branch and how we help people do things at scale. What are you seeing from your customers when it comes to managing these branch locations and all the complexity that comes with that?
Yeah, complexity. The good thing is with Cisco, you guys kind of remove a lot of the complexity. I do a lot with Meraki, with our customers. And observability and just visibility of everything across the branch is probably their number one request. Having everything in one dashboard, one visibility component is critical to what they want to do out of the branch office.
Amazing. So then you've seen all the announcements today. Are there any that you think, from your customer's perspective, you can't wait to get out there and show them that they're going to be excited about?
I think the buzz in the street is all about AI. With AI coming on strong with the AI agents, with the generative AI you have going on, with the AI Canvas, with the unified dashboards, it's going to be a lot to unpack. Really having another way to interface with your dashboards, with your devices, just takes the troubleshooting and effort to the next level.
Yep, amazing. I think the fact that we're unifying a lot of that experience, bringing it all to a dashboard, we're hoping that power of the existing portfolios, the simplicity of the Meraki portfolio, we're really hoping that lands.
Oh, by far, yeah. A lot of our customers are on the Catalyst environment. Our involvement is to really try to get them into the Meraki dashboard from the hybrid operating mode and the monitor operations, just to make the best of really both worlds. It's been really exciting for us to test out.
You don't have to tell the bad stories. I want some success stories. Any examples of some great things customers have done with the technologies where Cisco and Ingram Micro and the customers have partnered together to do great things?
Yeah. I mean, it's really the ability for us in distribution is to get them excited for more of the technology. We started with these are basic conversations with wireless. We get in the next conversation of adding the switching to it. Eventually we understand what's the outcome you're trying to really drive. Where we unpack it beyond the technology and say, OK, you're trying to solve for this. We really start exposing into the Cisco marketplace, the ecosystem partners. It just really solidifies them as the value add in that reseller partner to their end user. It really makes them just differentiated across the board.
Amazing. Anything you're personally excited about outside of the Ingram Micro lens?
For me, it's the security everywhere. With that being in all the platforms, the ability for kind of that next step of saying that we can really secure any point of the environment and then protect you for the future, that's going to be the exciting thing for me with really no questions asked.
Fantastic. Thank you, Dan. I appreciate that.
Thank you.
I'm going to turn it back over to Rob. Thank you.
I was actually take his microphone real quick. I was just going to ask you a quick question. Nice job, Daniel. Thank you so much.
Thanks, Dan.
Yeah, just real quick. You guys did make some really big announcements this week that I think are substantial changes, especially from the management front and the continued integration. Any priority thoughts we should have about where to pay attention to what you're most proud of and where we should be watching you guys?
Yeah. Look, it does start with the idea that security and the network are inseparable. The fact that we have a consistent way to apply security policy right across that network, I think, is massive. I think the customers will benefit from that. The network is going to be more secure, more consistent, more reliable in that way. Certainly, a lot of the devices we've launched we're very excited about, but they do not really sing without the great management on top of that. The idea of driving more of that to the cloud is something we're invested in very heavily as well.
Love the work you guys are doing. Congratulations. Nice job on the presentation here. Guys, be sure and check this stuff out if you get a chance. Steve, we'll go back to you in the studio.
Rob, thank you so much, brother. Our thanks go out again to Lee and to Daniel. Fantastic job. Great, great, great session. We are about a minute and 20 seconds from our next keynote deep dive. This time, we're going to focus on how AI changes everything. We're hearing all about it this week. This is a new blueprint for network security and zero trust and for the SOC. We have got Tom Gillis on deck, Senior VP and General Manager, Infrastructure and Security Group, along with Mike Ford, Senior Manager and GP of Splunk Security Product. What they're going to do is give us an exclusive deep dive into smarter, more effective security solutions, all of the different ways that Cisco is helping to redefine security everywhere that the network meets directly with AI. Cisco has got decades of experience in networking and in security.
We're delivering our most significant security innovations in literally 40 years, all of it powered by AI. So exciting. Let's talk very, very closely about what we're going to hear about hybrid workforce security. We're going to look at firewall disruption over the next few minutes. They're going to do a bit of a deep dive into more enhanced security analytics, the way Cisco and Splunk are working together to revolutionize security. Customer success stories. We're going to hear from a customer, Cleveland Clinic Foundation. Very exciting. Right now, we are going to send it out to Tom and Mike. We'll meet you right back here in the studio on the flip side. Away we go.
Before we bring on Tom, let's check out this video.
All right. Good afternoon, everyone. I'm Tom Gillis. I'm the General Manager for Infrastructure and Security here at Cisco. We got a full house here, which is great. Can I just do a little sound check? Can you hear me in the back OK? Thumbs up? All right, good. I'm going to talk about some of the things we're doing in security, some of the trends that we're seeing, some of the interesting observations. Maybe we can make it a little bit interactive if possible. The way we think about networking in general and infrastructure at Cisco is three big buckets we talked about this morning: AI-ready data centers, future-proof workplace, and then digital resilience to tie it all together. Security is infused into each one of these areas.
During this presentation, we'll talk about how security can be built into the infrastructure in a data center to protect AI-based workloads in ways that really were not possible even two years ago. Workloads do not just live in the data center. In this AI world, workloads are going to be everywhere. In the workplace, there are all these robotic AI-powered devices out in a hospital floor, in a factory floor, in a campus. We can put this out in the future-proof workplace. The digital resilience is the analytics that can tie it all together. I have been doing this security thing for a long time. Every time I come to a security show, you hear the same story, which is the threats are getting worse. It is true. It has been consistently increasing in severity and threats.
I would argue that what we've seen in the last year is something that is a bit new. We have seen attacks that are aimed against switches, routers, network infrastructure. They're aimed at our critical systems: power supplies, water systems, telephony. You might not know it, but some of the switches and routers in your environment. The motivation of these attackers is different than what we've seen before. The motivation is to penetrate, get in, and stay in so that at the right moment, they can turn the lights out. This is a situation that we have to take very, very seriously. In the post-AI world, the nature of what we secure is changing so dramatically that security needs to change in ways it hasn't really done in the past decade.
One of the most fundamental changes is the nature of the application itself is completely changing. We used to think about an app in terms of three layers. There was a presentation layer, an app layer, and a data layer. Now, as we begin to AI-enable these applications, there is a new layer. It is called the model. The model sits between the data layer and the application layer. The model ingests all your data. Did you hear the keynote this morning where Kevin from OpenAI was talking about the more data and the more compute we put at these models, the smarter the models become? That is a very good thing. Your model is going to learn all your secrets. What you may not realize is that when a model learns something, it never forgets.
You can't go in and delete a record from a database like you can in a database. When a model learns something, it gets kind of infused into the model, and it never forgets. The way that we think about protecting that data, securing that data, securing the application, it's going to fundamentally change. You've heard us talk about our platform approach. In security, we think about the security platform at Cisco in terms of three major platforms. We have our user protection suite, which is all the stuff you need to protect a user, the cloud protection suite, which protects applications, whether they're traditional apps or these AI apps in a cloud, and then the security and analytics and response to tie those things together.
We're going to go through and talk about each one of these three suites and some of the new developments that are happening there. OK, we're going to start with an anchor of the user protection suite, which is zero trust network access. The idea behind zero trust network access is actually quite simple. You want to implement least privileged access for your users. A user, a salesperson can get to sales apps. An IT person can get to IT apps. You don't want salespeople getting to IT apps ever. You want to be able to apply those policies not just to users that are remote, but you want those same policies when a user is on campus. You want to have the ability for remote users and campus to have the same policy.
You want those policies to apply not just to users, but to things. I often say people forget that printers are people too. That printer in your remote office, it needs to access the print manager, or your telephones need to access the call center, and only the call center. You do not want a printer accessing a customer database. This is where traditional zero trust really struggles, but Cisco really shines. In a traditional approach to zero trust, you authenticate, and you check the posture of the machine when it is authenticating. Then you allow it in. Once you are in, you are in. You can do whatever you want. With Cisco's approach, we are using continuous risk assessment. We do not just authenticate you at the, we do not just assess you at the time you authenticate.
We're continually looking at, has something changed on your machine? If a host-based firewall setting gets changed, then the security posture goes down. It is a continuous assessment of your machine. It works the same for users as it does for devices. This is Cisco's great strength. Using security group tags and ICE tags, Identity Services Engine, we can identify that printer or that telephone at the point that it connects to a network because a printer is not going to log in to a web proxy and identify itself. This is something that customers are increasingly interested in: how can I deploy zero trust solutions for users and for things?
Now, there's another nuance in here, which is if you have modern applications that fit into a zero trust framework quite nicely because they can tolerate a web proxy or gateway sitting between the user and the client. Many of you have many older applications that were built in the original network model where the user goes directly to the application. When you stick a proxy in the middle, it can break. It breaks for a whole bunch of reasons. You know the feeling. It breaks if the app is expecting to see the IP address of the client, server-side initiation. It'll break if it's peer-to-peer. It'll break if it's multichannel. The most common and my favorite is like, I don't know, it just broke because nobody ever tested it with a proxy.
One of the unique things about the Cisco solution is we have a single client that, if it needs a zero trust solution using HTTP, we'll make that connection. If it needs an IPSec connection using VPN, we'll open an IPSec connection. The user doesn't know we're there. My tagline is, the company that bought you the VPN is killing the VPN. We're not actually killing the VPN. We just take it out of the path of the user. This single client provides seamless connectivity for legacy apps as well as new apps for users that are remote or on campus. We do it in a super efficient way. All of this has what I call the magic of Duo that is built inside. Duo has added identity. We are an identity client that federates with your identity store.
We're what's called an identity broker. We store a subset of the data that is in your master identity store. This is where we do the authentication. What this allows is, if you have temporary workers, if you have contractors, you can put them into the identity store in the secure access. You're not going to pollute your Active Directory if that contractor is only on for weeks or days. Just a tactical solution that makes it much, much easier to manage identity at scale. Continuous risk assessment, focusing on delivering a great end user experience wherever that user is. One of the things that reflects the Cisco platform effect is we measure that end user experience. We integrated the ThousandEyes client into this secure client access. If the user experience isn't great, we can tell you exactly why.
Is it a problem with their laptop? Is it a problem with their browser? Is it a problem with their Wi-Fi at home? Is it a cable cut deeper in the network? Is there an outage in the data center? We give you an end-to-end view. This is, again, unique to Cisco and ThousandEyes. Continuous risk assessment can make a smooth, seamless end user experience wherever that user is, whether they're remote or on campus. Now, more and more users are using apps like ChatGPT and prompt-based AI. We've been investing very heavily in building AI security capability that you can use as a point solution. What's much more interesting, in my opinion, is we've built this as a feature into secure access.
If your users are going to a prompt-based interface, we can understand the interaction with them and look for compliance, for data leakage, and simple visibility into who's doing what with prompts. That's just a feature built into Cisco Secure Access. Now, we further integrate Secure Access into our networking offering. This is what the industry calls SASE. One of the pieces of news is that with Meraki, we've now done an integration where Secure Access is built into Meraki. The AutoVPN function will reach Secure Access. You can provide this on Catalyst or Meraki platforms. Really, the focus here is to deliver an integrated solution that can leverage the power of an SD-WAN solution, the tagging of Identity Services Engine and ICE, and the Cisco Secure Access, all in one integrated package that is easier for you to deploy.
To talk a little bit more about real-life experience with zero trust, I have a special guest that's going to join us, Shawn Clark from Cleveland Clinic. Shawn.
Thanks, Tom.
Thanks for coming.
Yeah, my pleasure.
OK, Shawn, we have been working together for the past year looking at zero trust. I find that least privileged access is particularly challenging in a medical environment because identity is so complex. Can you talk a little bit about the clinic, the size and scale, and what some of the challenges are there?
Yeah, absolutely. Cleveland Clinic, based in Cleveland, Ohio, we have about 85,000 caregivers spread along 23 hospitals, about 280 clinics that cover four countries and three continents. Network goes between all of that. We have somewhere around 1.2 million devices on that network.
The reason why least privilege is challenging is because you have complex roles and relationships. You've got to have a doctor that is associated with a hospital but is also a researcher, say. Right?
Right. Identity is really one of our big cornerstone issues. The doctors are a great example of why, but certainly not the only example within our environment. When you have a doctor who's in clinic one day, the next day is interacting with the executive board because he's an executive medical officer, the next day he is working as a researcher, and then the following day, he's at home trying to catch up on his pajama time to get all that paperwork and stuff done. Tracking that identity and providing a frictionless access to the different data he needs or she needs is absolutely critical.
Yeah. You have a complex set of identities. The systems that they interact with are also very heterogeneous and complex. This is an area that we've worked together to build some integrations there. Can you give us some examples of what we've done together?
There are certainly a lot of examples of that as they work with different systems throughout all those different personas. One of the great examples, though, is when they're in the clinic and working with badging in and badging out as they go between all of the different clinical exam rooms, working with patients. The amount of time it takes to do that, to move between the different exams, and to make sure that it's not just frictionless but that they have access to patient data, they have access to maybe their executive data, et cetera, is quite challenging.
Yeah, yeah. We worked together integration on those tags with a company called Imprivata. I remember you telling a story. I think it was in a different context. The difference that one click versus five clicks or 10 seconds, a different 10 seconds makes when it's repeated over and over. Can you share?
Yeah. Before coming to Cleveland Clinic, I worked in South Florida at the Bascom Palmer Eye Institute as a PACS administrator. After an upgrade to our PACS system, I was rounding through the clinics, making sure everything was OK. One of the top retinal surgeons in the world saw me from down the hall. He goes, "You, come here in that room right now." As I go in there, he's like, "You added one click to my workflow." He then proceeded to tell me how he had calculated out that one click meant that he did not see four patients by the end of the year. That is one thing for one doctor. When you scale that up to the size of Cleveland Clinic, now you are not seeing a lot of patients.
You're not just impacting patient access and patient care, but the health of the community as well because you're talking about that many patients.
Yeah. Now, if you listen to my opening commentary, I talked about least privilege for users as well as for machines. You've got a lot of machines in your environment. Right?
Oh, we have an amazing network. We've got everything from Xboxes in our children's hospital through all of our clinical workstations, all the way up to a quantum computer on campus.
Right. And so having the ability to apply tags and use the network to transmit identity, foundational. Right?
Absolutely. Because with all of those different devices on there, we have to be able to push our policies and our control points as close to our devices as we can. SGTs are one of the ways that we do that. That integration with ICE and secure access is a critical part of that.
Now, the other thing I like about meeting Shawn and this experience is that I've always had a kind of litmus test for delivering a great user experience with great performance. That litmus test is making the radiologist happy. The reason why we laugh is that radiology, they have to handle big files, like gigabytes.
Five gigs or sometimes more. Yeah.
They're interacting with them. They're going through and looking at the X-ray of the imaging. They're doing it from home. These guys and every doctor thinks they're an IT specialist too. Right?
Oh, yeah. Yeah. Radiologists tend to be very technical.
We have been very focused with secure access on using the latest protocols. We use a protocol called QUIC, which is a streaming protocol. It delivers a very meaningful performance advantage for these large files. Are we going to make the radiologists happy together?
We are, actually. We are already making the radiologists happy just in very early testing of how to really optimize their connections from home, which are varied. Some are consumer grade. Some are commercial grade. Sometimes they have Wi-Fi access points. It's very heterogeneous.
Yeah. Performance really, really, really matters in delivering that great user experience. We'll do this together in the wild. The other thing I wanted to chat with you about is, as we think about streamlining and automating infrastructure broadly, firewalls are particularly problematic to upgrade. Can you share with everyone the firewall upgrade story?
Yeah. Last year, we were able to replace some of our aging ASAs with FTD 3130s. That process was one that took a very long time to get to the point of doing that. When we finally did the cutover, we had some of our team going through the hospital emergency rooms and, hey, how's everything going? Is everything all right? One of our team members walked up to one of the doctors in the emergency room and said, hey, we just changed the firewalls. Did you see anything on the internet? He said, you did what? He goes, no, I don't have time for that. I didn't see anything. Just leave me alone.
That's how we want it to be, invisible. Right? It just works. Last thing we're going to touch on, as we think about this world of AI and the speed at which it's all moving, do you see agents operating in the hospital or when I say operating, running in the hospital or in your environment? Is this a concern?
Yeah, it's a huge concern. It's not just a concern from a doctor being in the clinic, but everybody within the hospital is using that, including the IT department. Right? So being able to understand where it's coming from, who's using it, what data they're putting in there, how it's impacting the network is absolutely critical to us.
Shawn, it's been a pleasure. Thanks so much for joining us.
Thank you very much.
We'll keep the radiologists happy. OK. Now I want to talk, we talked about the user protection suite and zero trust access. Now I want to talk about the cloud protection suite and what we're doing with firewall. The industry has a term called hybrid mesh firewall, which is an evolution of where firewalls are today. The idea is you have multiple different form factors that allow you to put firewalls in lots and lots of places. This is something at Cisco we think we can do particularly well because we really want to be able to take network security—I won't even use the word firewall—network security and break it literally into a million pieces. Imagine a mesh of a million little tiny firewalls. Now, these firewalls could live in a big hardware appliance like you know and love today.
They could live in a VM of that appliance. They could also live in the host itself using eBPF, which is a capability that's built into Linux. We'll talk more about that in a different form. It's a very, very powerful capability. The most interesting thing that we've announced recently is we're putting hardware accelerators into switches and routers that allow this little baby firewall to live in a switch itself. That's what's called a smart switch. This means that every port can be an individually configured high-performance layer 7 firewall. It can also live in a third-party firewall. The idea here is that we're not going to ask you to rip and replace existing firewalls. Rather, we want to meet you where you are. You can age them out.
The news that we have is we've introduced both a low-end firewall and high-end firewall. The 200 series is a 1.5 gigabits per second firewall designed for branch offices. At the high end, with 6100, we'll do 400 gigs of firewall plus IPS or 200 gigs per rack unit. Very, very cost-effective, very dense performance form factor. This really rounds out our whole line of firewalls. We've refreshed all of the hardware firewall. We believe that we are a significant price performance leader in firewall across the board. Now, the heart of the Cisco hybrid mesh firewall is the security cloud control. It's the one controller where your policies live all the time. The enforcement points underneath that can change over time. The policy doesn't. What we've announced, there's two new products that we've announced.
We've introduced the secure router, where we're putting firewalls into a Catalyst switch and router. We've also done an integration into Cisco ACI. From this one policy console, you can push your policies into multiple different enforcement points, including all these different Cisco devices as well as these third-party firewalls. This is what allows us to meet you where you are, where you can deploy the Cisco hybrid mesh firewall one workload at a time, little tiny bits and pieces. You don't have to do some giant forklift of ripping out existing firewalls. It's a smooth and simple transition. Now, as we think about that platform effect, we're doing some really interesting things to integrate this firewall into Splunk. I'm going to ask my colleague and friend Mike Horn to come up and join. Mike is the GM for Splunk Security.
Hey, Tom.
Good to see you.
How are you?
Thanks for coming. Mike, you and I have been working around firewalls for a long time. In the context of Splunk, the firewall is a critical source of data. Right?
Absolutely. Probably one of our top two sources of data in Splunk for a lot of folks.
It is generally one of the bigger sources of data, right?
That's right. How many Splunk customers do we have out there? Just show of hands. Right? Pretty good.
Yeah, pretty good. We want more.
Yeah.
We are about to make that easier, Tom. For those that did not hear in this morning's keynote, one of the exciting things that we wanted to announce is that we are providing security analytics for Cisco firewalls in Splunk at no charge. There is a new program coming out, you can see, in August this year, where you will be able to bring Cisco firewall data into Splunk and get the analytics, the detections, the visualizations, all those great things as part of being a customer of Cisco and Splunk. This is a big deal. Right? For many customers, firewalls can be a third of the total ingest. We will not charge for it. As we continue to work together on our vision for bringing Cisco and Splunk closer together, we will be able to federate with Splunk's distributed search capability directly into the firewall log console.
In the future, you won't even ingest at all.
Yeah. It's super exciting. Lots of good things coming on that.
Now, the nature of the firewall is also changing with these smart switches. This applies to the same program. Right? Having the ability to inject a firewall into a network device, you can think of this as one device that has two personalities. One personality is around packet forwarding. That packet forwarding, you want simplicity, stability. You want this thing to run for years without a reboot. In fact, many customers have Cisco switches. Remember the 6500? That can run for a decade without a reboot. On the other personality, it's about security. Security wants to update.
Yeah, as you said, all the time because of the changing threat landscape, for sure.
Right. What we've brought together is the security is processed in the DPU. This is lifecycle managed, independent of the switch, which is super important. The security team has an interface where they can set up their security policies. They can update the firewall code. It doesn't touch or modify the switch. Two personalities living in one box. We think this is going to be a huge part of your security arsenal going forward. The idea here is that the network processor is free to do what it does, which is move packets from A to B as fast as possible. This advanced processing that we do with the DPU will help us identify friend from foe. Is this packet legit or not legit?
That telemetry, we continue to find new ways to bring that into Splunk for additional insight.
Absolutely. Tom, what do you love to say? Where do you find and see lateral movement?
By definition, on the network. Right? I think this is Mike and I, our mission is to continue to unlock east-west traffic so the combination of Cisco plus Splunk can identify lateral movement better than anyone else. Really, who else can do this?
Yeah. Yeah.
There's another more practical aspect to the value of a DPU-based switch. When you're injecting a firewall into a network infrastructure, you typically use two switches, two firewall appliances, two more switches. That's six boxes. When the firewall's built in, six boxes becomes two boxes. Oh, and by the way, they upgrade themselves. Right? Operational simplicity, lower cost, just less cables, less power, less stuff to think about. We think the value proposition of the smart switch is really, really strong. It's going to really transform the industry. I talked in the beginning about the nature of these threats. You can read about this in the paper. Volt Typhoon, Salt Typhoon, these are very, very serious challenges and something that we've spent a lot of time looking at.
One of the learnings from Volt Typhoon and Salt Typhoon is that when you look at both applications but also the infrastructure itself, you've got to recognize patching is hard. This is something you and I have been looking at for a long time. Right? How often are vulnerabilities found in the enterprise?
Yeah, exactly. I mean, in the speed that a vulnerability gets disclosed to when we see active exploits in the wild, it used to be weeks, months. There was a pretty long lag time between a new vulnerability. Sometimes it took years. Now we're seeing that get shrunk down to hours. Some of that's based on AI and other capabilities that people have. The fact of the matter is these exploits need to be patched. The vulnerabilities need to be patched much quicker.
Yeah, with some sort of an automated mechanism. Right? You can't patch all of your infrastructure in hours. It's just not reasonable. The other thing that we've been observing, and Mike, I think you have too, is that there's been a big drop-off in zero days. Have you guys noticed this? Like less fire drills, less panics. We're like, oh my god, the new zero day. The reason for this is that attackers are realizing, I don't have to find a zero day. That's hard. Right? We still see activity around Log4j five years later. Right? Half a decade later. You see the same thing?
Absolutely. Yeah.
Yeah. Existing vulnerabilities are a pathway in. One of the areas that we've been focusing on with this distributed system is, can we take the vulnerabilities that we know about in your application and automatically apply a compensating control? Now, a compensating control is not a patch. It's a shield that can protect that vulnerability.
Yeah. What's different here than other patching and virtual patching that you've seen before?
Yes. It's a good question because the industry, Mike, because we've been talking about virtual patching for like a decade. What's different about this is that with a distributed system, with the granularity that I talked about with Hypershield, where each switch port can be its own little firewall, each VM gets its own little firewall, you can apply the compensating control to that particular service of the application. So web service, a database, a Postgres server with a vulnerability, we know, oh, this is a vulnerability from Postgres. We apply that compensating control right under that individual service. That's really the key to it is this very, very distributed nature and the automated management of all this stuff. Right? When the vulnerability is updated, the compensating control is automatically removed.
The hybrid mesh firewall at Cisco is a mesh, which means it can live simultaneously in the operating system using eBPF in Linux. That is also now coming in Windows. We have that available in Windows. It can live in a DPU in a server, or it can live in a DPU in a switch and provide those compensating controls for your applications. Patching an application is hard. Mike, is patching a switch easier?
Yeah. Did we just talk about people want switches to run for how long?
A decade.
Yeah.
Yeah. Right? That creates a bit of a security issue. One of the things that we announced today, which is a pretty big deal. System in the switch without rebooting the switch.
This is really cool.
It's important to note that a compensating control does not eliminate the need to patch. It is a finger in the dyke, right? It gives you the ability to patch in an orderly fashion. It is not going to be a crisis. We think this is a really big deal. We think this is going to change the way that you operationalize your network. What we are moving towards is a level of autonomy and kind of dynamic updates where this stuff is going to keep itself up to date and protect itself. I think it is necessary in this AI world.
Absolutely.
As we think about this AI world, we need to think about how we can understand the behavior of that model in an AI-based application. Here's the way I think about it. We want to think about tools that can help us validate and protect those models, both before you put them out and also during runtime. If you remember, the model is learning all your secrets. The app team will put guardrails in place to prevent the model from talking about certain things. What we're finding is that an AI-based application is nondeterministic, which means you can kind of trick it. The way I think about it is, when you were kids, do you remember the game 20 Questions? OK? I've got a secret. Mike, you got 20 questions to guess my secret. I have a secret, Mike.
You're going to guess my secret. What is it?
Tom, I already know. You're thinking about your 20 questions.
You got 20 questions in your number right now. I don't need one.
He knows. That's not fair. The idea here is that with AI, with AI Defense, we're using AI to protect AI. We have automated the process of red teaming the model where AI Defense will play the game of 20 questions a billion times over. It'll constantly keep asking the model and looking for vulnerabilities to see if the model will reveal your secrets. This is super, super important because at the speed at which things are moving, I guarantee you in your environment, people are downloading new models all the time. There's so much innovation happening with DeepSeek and this model and that model, Llama. Right? People want to explore these things. It's just not possible to red team and verify the model manually. You need an automated solution like this. This is where we really shine.
We do it not just before you deploy the model, but if your developers say, OK, we're ready to put this model into production, we can do this by being able to enforce those guardrails into the fabric of the network. Again, thinking about that platform effect, the AI defense is being built into the hybrid mesh firewall so that we can provide those guardrails that are independent of the application. Whatever the app team does, you can guarantee that your security guardrails are going to be met even in this post-AI world. Mike, we've been working hard on bringing Cisco and Splunk better together.
Yeah. One of the things I'm really excited about is this was the first new Cisco security application that had native Splunk support on day one. We were able to bring that support in.
We can take alerts out of AI Defense today and action those inside of enterprise security. In future versions, we're going to have even more telemetry coming out of AI Defense. You can write your own detections. It depends on the things that you're looking for on how people are using AI models. You might have keywords. You might have certain things that are specific to your environment. You want to make sure that you're protected against those or you're getting detections when those things are happening. We're really excited about what the future holds here.
Now, all of this stuff is managed from Security Cloud Control. This is the single console that allows us to do hybrid mesh firewall, the third-party firewalls, AI defense. It's all built into one interface. The idea here is, I said it earlier, for many of you at enterprise scale, this could be a lot of enforcement points, like a lot. Right? Hundreds of thousands or maybe even millions of enforcement points. If you're familiar with firewalls, you might be thinking, I can't manage 100 firewalls, let alone 100,000 or a million firewalls. We really have to have a new management model. With the power of AI, what we are building is a firewall that literally writes its own rules, tests its own rules, deploys its own rules, and upgrades itself. Let me explain how this works.
With this distributed hybrid mesh firewall, for each instance of the firewall, we actually run two. It's called a dual data path. We can think about the software upgrade process. When you're running version 2.0 of the firewall, that goes in the primary data path. When 2.1 becomes available, we put that in a shadow data path. There's a local AI engine that compares the two. Not in a sample of your environment, not in a staging area. In all of your environment, you now have a digital twin. Think about it. Policy changes. When you push a policy change, you push it into the digital twin. When you do that compensating control, you push it into the digital twin and see, is this going to do something unexpected? Yeah. Big deal?
Yeah. Totally changes the game in terms of being able to, like you said, hey, have this thing self-develop its policies. But how do I know that they're working? Well, I can apply it in that shadow data path. I can see whether or not all the things that are expected to be connected are able to connect. And then I can apply that automatically. So huge, huge change.
The reason we're so excited about this is that as we develop our hybrid mesh firewall, we're going to be able to provide better protection for your applications. We're going to have access to much more fine-grained visibility. We're finding these efficient mechanisms to deliver that into Splunk so that your security team can take action with advanced AI-powered security tools and identify these types of threats. To talk more about that, I'm going to invite one more special guest up here, Peter. Peter Bailey has just joined Cisco. Yes. Thunderous round of applause for Peter Bailey.
New guy.
Peter.
People, vision, impact. Just really quickly, I think you all know the people at Cisco are incredible. Great culture, very professional, dedication to the mission. We all get to choose who we work with. That was an easy one for me. Vision, I think the vision you saw this morning and the vision you're hearing today is just so compelling. We're at this moment of, like we were back in the 1990s, of rebuilding infrastructure for this new era, this Gen AI era. Security needs to be infused into that for it to be effective. As a cybersecurity nerd, I mean, I can't imagine a more cool opportunity to be working on that problem when that gets to impact. You want to impact not just one customer, but thousands or millions of customers. For me, no brainer. Sign me up. I'm in.
Let's go.
I'm not going away. But I'm focusing on where security meets infrastructure. The torch is officially passed. I'm going to say goodbye to you two guys and let you finish up. OK?
Thanks.
Thanks, Tom. Always a tough act to follow when you're following Mr. Tom Gillis.
Totally. Totally. I've got to come up with better jokes, I think.
Yeah. Yeah. For sure. We are going to talk a little bit about security operations and analytics and the things that we see. Peter, I know you are out talking. You have talked to lots of customers in your former role and in your current role. Maybe you talk a little bit about what you are hearing in terms of key trends and themes.
Yeah. I spent a lot of time talking to CISOs and their teams. Listen, you heard the story. Threats are still increasing. They're getting much more sophisticated. I think the other factor you hear a lot about is that the amount of data we're now tracking and trying to manage and analyze is just exploding. Data volumes are going up 30%-40% a year. It's getting more and more arduous and expensive to bring all that data back and then analyze it and make sense of it at the core. Really this idea of, can we go to the data? Right? Is there a distributed data strategy we can engage in, which also addresses costs and time and all that? Can you do that but still have centralized best practices and workflows to protect your operations?
Those are two really common themes I hear.
Yeah. I hear that a lot. The fact that, hey, I've got my data spread out across a variety of different data stores. We'll talk about that. I need my analysts to be able to work in one single—how long have we talked about single pane of glass?
Yep.
I think we're getting a lot closer. I know that that's a vision that people have for how they want their SOCs to operate, particularly when we talk about AI. If I look at this evolution in the distribution of data, we used to say, bring it all to one data store. Bring everything to one place. That was certainly part of the Splunk of an era ago. Now we've evolved and said, look, there's a lot of different data locations. We have what we call data lakes, data ponds, data puddles. Right? You have all this data. A lot of times that data has gravity. There's a reason that you want to keep it in the location where it originated. That might be it came out of a high-volume cloud service. It makes sense to keep it in that cloud provider.
Could be that it's a small little piece of data that's sitting in some application. You just want to leave it there. We see that customers have a need to be able to access that data where it lives across all those things. They really want to do two key things. Right? The first is they want to be able to search across that data. We call this federated search. The ability to say, from a Splunk interface, I want to search for an IOC. It might be an IP address. It might be something else that you are using as part of an investigation. I want to search for that across the data no matter where it lives. That's the first piece. Second piece in security, we're always on the lookout for what? We're looking out for threats.
Detection.
We want to be able to detect and see, is there any evidence of a potential threat in my environment, again, across all these different data stores? At Splunk, we've come up with a solution that we call federation. At the heart of that federated solution is an updated data management set of capabilities. If you think about all the different sources that are generating data, cloud, endpoints, OT, IoT, we heard earlier from Sean about 1.2 million, I think it was, Sean, devices on your network, all generating information.
All the legacy applications too, right?
Yes.
Being able to go to those and grab that and analyzing in its existing format.
That's right.
I mean, super powerful.
That's right. You have all this data. You want to be able to groom that data. You want to be able to filter it. Right? You want to be able to redact potential PII. I'm sure that's a thing that you care about. You want to be able to route it to different data stores based on the use cases. Hey, this is compliance-only data. I want to route it to a data store where it's just mostly sitting. It's there if I need it, if an auditor comes and I need to show that I've got it. You want to be able to do those things that I was mentioning, which is being able to search and detect across all that. We call that in Splunk federation. We have federated search and federated analytics.
Lots of exciting things that are happening there. That is how we're helping customers address the distributed piece of the key theme.
Yeah. Just on that, again, have worked with lots of multinationals. Lots of companies have done lots of acquisitions where everything's all over the place. Really, really powerful feature to be able to offer, a capability to be able to offer. I think it ultimately also is going to have a lot of economic benefits too because you're going to not have to pull everything back and not have to not only ingest it and all the egress costs and all that, but also all the compute to normalize it, analyze it, and all of that. Super, super cool.
That's right. The other thing we talked about was this need for a single pane of glass, seamless enterprise or analyst experiences. What are we doing there?
Yeah. Listen, I think in the security operations game, you want to try to unify workflows, drive best practices, drive repeatability. The more you can unify those workflows, the better results you're going to get out of them. Really, it's about trying to unify threat detection in the SOC and then ultimately drive automation.
Yeah. Yeah. Absolutely. AI playing a huge role in how we accelerate all that. Let's maybe talk about one of the products in the portfolio that helps do this.
Yeah. Cisco introduced just an incredibly innovative product last year called XDR. Think about it as a detection layer as well as an automation layer for doing investigation and ultimately response. This allows SOC teams to operate at speed, especially smaller teams that may not have the resources to do the investigation and response work. Super cool. It's just the beginning. If you think about being able to automate the entire process and ultimately doing remediation across that process, that's the kind of tools we need to be providing when threat actors are acting at AI speed and in much more sophisticated manners. It's the kind of stuff we're doing to innovate that I think is really exciting.
Yeah. Tight integration with things like the network data sources and being able to get network telemetry, looking for that lateral movement. You can take an alert that came out of XDR or any of the other security tools that you have and leverage Splunk's enterprise security to do further investigation, to do deeper analytics, be able to write your own custom detections if that's something you need, be able to automate a lot of that workflow in terms of, hey, where else have I seen this, being able to look for patterns and trends and all sorts of different things. We've got an AI capability as part of that that we call an AI assistant. That'll run activities, things like being able to get an alert that's getting summarized. Hey, help me understand this threat in the context that I'm looking at it.
Oh, there's five threats that are all related. Being able to do natural language queries. Being able to say, show me all the servers that connected to the C2 IP address rather than having to figure out and write the SPL that's required to get that information. That's a huge game changer. One of the things I was talking to Sean about this a little earlier, it may not be quite as sexy, but generating the reports of, here's all the investigation activities that were performed. Here's who did that. Here's what they found. Here's the supporting evidence for whatever the finding was, the determination, a false positive, true positive, et cetera, saves teams tons of time. The people I know, they don't hire security analysts for their report writing skills. They hire them for their creativity, their inquisitiveness, et cetera.
It's going to be a game changer on that front.
Yeah. Absolutely. Bringing it all together, if you think about what we're doing at the infrastructure layer, what we're doing in terms of firewall, next generation capabilities, all the way up to the analytics stack. I know when we asked the question before, who's a Splunk customer, we did not get a ton of hands in the room because this is more of a network audience. I get that. What you are going to see more and more is that architectures you are deploying in your organization from a network perspective are going to include security considerations. These things really do need to come together. We have been solving these things actually in the network. Having Splunk on top of that, I mean, it is really the best in class analytics layer. Splunk has really defined this industry and continues to innovate across the top of it.
It is really just bringing it all together. I think that is really the vision that is so compelling that Cisco is pursuing. Super cool stuff.
Yeah. I'm super excited about how all these pieces. You heard, you saw in the keynote earlier about some of the integrations. We've done a lot of work in terms of bringing Cisco data into Splunk and making that a first class experience. Lots of things happening in terms of threat sharing and other things. Talos is now integrated with all the Splunk security products. You get the benefit of Talos security information into your Splunk solutions. We're super excited about this. I think the rallying cry is, hey, let's go build the SOC of the future together.
Yeah. Absolutely. Absolutely. I think we got maybe one or two more slides here because we have to make announcement about Tom's next thing too. Kind of as a closing comment, I mean, how cool is it? We're putting AI defense in the firewall. We're putting the firewall in the switch. We're putting the switch in the compute infrastructure that's going to power GenAI. You have an analytics layer across all that for observability as well as security to really analyze, protect, and defend that environment. Again, to be good at AI in this new world, you have to be good at security. It is absolutely critical. What we haven't seen yet are also the fact that threat actors are going to start using these tools in ways that are far more advanced in terms of prevalence, in terms of sophistication of attacks.
It is also critical that we do make these investments. We got a shot at doing this right, introducing GenAI in a secure way. Cisco is committed to make sure that that happens. I am really, really excited to be here part of the team and also excited to work with all of you.
Great.
I think closing is one more.
We got Tom.
Yeah. Yeah. I mean, a quick plug is tomorrow at 2:30. Beyond that, thank you. We really appreciate your attendance here and enjoy the rest of the show.
Yeah. Thanks, everybody.
Tools at their fingertips, like real-time call summaries that can be shared when transferring a call. With the call sentiment for supervisors feature, supervisors can assist team members whenever they need support. Enable your local and regional employees to deliver seamless customer assistance empowered with AI in the collaboration app they already use. Webex Calling Customer Assist.
Cisco and SAP have agreed on ambitious milestones for the migration of our more than 250 offices globally. Right now, we are planning on having upwards of 350,000 devices on the network, including IoT devices. We have more than 100,000 employees. They need best-in-class IT infrastructure to work wherever, whenever, and however they choose with the devices they prefer. With Cisco, we've made our network simpler and more secure. Cisco SDA also enabled us to improve our Zero Trust Access policy, ensuring only trusted devices can access corporate resources. With Cisco Spaces, we could leverage this foundation to establish a global platform for smart offices. Our first experience confirms our hopes and expectations that we can manage everything in one place together with a professional partner like Cisco.
Tribeca has always been curious about technology. I feel that so much of Webex and Cisco specifically have done for us is help us lower the barrier of entry so that all artists can participate and be involved in the festival.
We're using Webex events in order to make the experience of De NiroCon much more interesting, immersive, and connected for our audience. A frictionless experience was key. We don't want our fans and customers worrying about how to find things or where to find things. All we want them to worry about is how do they have the best possible experience. We want to continue to tap into the broader suite of capabilities that Webex has, including a lot of the AI capabilities that they continue to build out.
It's important to work with companies that have shared values because it's a partnership. It helps us amplify our voice. To a certain extent, it's a shared mission of Cisco to be able to amplify diverse storytelling. You need to have global partners if you're a global festival. We're truly honored and proud to be partnered with Cisco.
Meet AI receptionist for Webex Calling. Your always-on virtual receptionists that can answer questions, book appointments, and transfer calls so your business can boost customer satisfaction without adding staff.
Thanks for calling Florissa Dental. I'm your AI receptionist. How can I help you today?
Hi. Yes, I have a toothache, and I'd like to schedule an appointment. First, I want to make sure that you accept US insurance.
Yes, we do. U.S. insurance is covered 100%. Would you like to speak to an insurance specialist, or are you ready to schedule an appointment?
I'd like to go ahead and schedule an appointment. Thank you.
Great. I'll transfer you to a nurse who will give you an initial consultation and help to schedule your appointment.
Thank you.
Deliver outstanding customer service. AI receptionist, available later this year as an add-on to Webex Calling. Take your business into the new era of customer experience with Webex AI Agent, which answers questions and fulfills customer needs, all while delivering personalized conversational experiences. With Webex AI Agent, you can engage your customers through the channels they prefer, using natural language and human inflections.
Sounds like a special occasion. Okay. Based on your description and our SOC, I would recommend daffodils.
Oh, I want a dozen, please.
Understood. A dozen. Coming right up. Okay. I think we have what you're asking for. I've sent it directly to your phone.
Even if your customer changes their mind mid-conversation.
I've got it. Miami. That's the destination.
Of course. I found several flights that fit your preference.
With reliable guardrails in place, the conversation stays within the AI Agent's capabilities.
Could you also book me a taxi, by chance, for when I land?
I'm sorry. Unfortunately, that is outside of my scope. Anything else I can help you with?
You can be sure that your customer's needs will be met, their problems resolved, and that all important transactions completed.
Let's get that payment arranged. I've sent a text message with the confirmation. Have a great day.
Welcome to the new era of customer experience with Webex AI Agent.
Thank you for calling. You're caller number 25. Your estimated wait time is 37 minutes.
Once again, we want to welcome you all back to the live broadcast streaming to you from the San Diego Convention Center at Cisco Live 2025. What an incredible, powerful day it has been. I'm Steve Moultrie. We just got the expert track on how and why AI changes everything. Such a huge theme of this event, kicking off with the keynote this morning. In this case, we looked at a new blueprint for security for Zero Trust for the SOC. Tom Gillis and Mike Horn gave us the deep dive into smarter, more effective security solutions. Specifically, they looked into hybrid workforce security, how Cisco is securing against identity-based threats and enabling safe AI adoption. They were able to look at firewall disruption, the way Cisco is transforming the firewall landscape with groundbreaking innovations. They looked at enhanced security analytics from Cisco and from Splunk.
We heard from Shawn Clark, leader of the Cybersecurity Technology Detection Program at Cleveland Clinic Foundation. Great, great guest on how all these innovations are transforming their security strategy and delivering really exceptional outcomes. Another fantastic keynote deep dive. Right now, I have got such a special treat for all of you. I do not know how we did it. Somehow, we managed to get Billy Hackenson into the studio with us, Cisco's VP of Marketing. Billy, welcome. Congratulations on an amazing event. I'm so delighted to have you here.
You're so kind. Thank you so much. I really appreciate it. We had a great kickoff today. Energy's high, and we're feeling good about the rest of the week.
It's been incredible. We never know. Going into each year, we always say it's the best year. It's the best year. You got to do it now, right?
Of course. Of course.
This year, I think I'm going to put it to you. Maybe it's the speed of development. Maybe it's the fact that everybody has finally said, "I guess we can't wait anymore. We got to dive right in," right?
Yeah. Yeah. I'm not biased, Tony. I think that this is the best year ever. I think there's watching the keynote today, and you started with Chuck and Jeetu segment in particular. Today was really about having the opportunity to share the speed of innovation that we're doing here. I think Jeetu and his team, Jeetu is 11 months into that role. I mean, the speed of which, the intentionality of which we're feeling, right? This velocity has picked up. I think you can say it's the AI chatter and the market. Foundationally, what they got across today is the network, the secure network you need to run that is Cisco. The launches and what we're able to announce and say what's available in the hands of our customers today or next month or in the next two months, just that speed.
I mean, I've been here four years. For the first time, really, we're at this place where the customers are feeling it in the room, right? I hope they did, right? That speed and that pace and that clip truly is remarkable from many of our sides.
Chuck, that's really how he opened. Of all the things he could have opened, he said, "We used to be Cisco networkers. We know networking. We've been around. But there's probably never been a time where the network is more important than it is today." We've talked about it for years. We've always said that. Now, I think we've reached that inflection point where all of our Cisco users around the world are saying, "Thank heavens I have the network because if I didn't have it, I wouldn't be able to deploy AI and future-proofed workplace capabilities and AI readiness in the same way," right?
Yeah. I was sitting next to one of my colleagues at the keynote this morning. We were leaning over to each other. She leans over and she goes, "It's almost like we have our swagger back," right? Just looking at the videos and looking at the story we're able to tell. I'm so excited because tomorrow's really going to be about actually hearing from customers. Yes, we had CVS Health today, and he was fantastic. We've got Ford tomorrow. We've got the University of California, San Diego tomorrow. Liz will be on stage with one of our key partners around the agentic side of what we're doing in customer experience. If today was around laying the foundation and framework for, "Okay, here's what we're doing," tomorrow's really about the proof, right? It's getting into the dirt.
Getting hands dirty.
Exactly. Showing what customers can do with it and where we can be the trusted partner along the way across the entire set of outcomes.
Absolutely. I want to kind of pick your brain a little bit because you're unique. We don't usually get to talk to people on the marketing side, how we tell that story on an outbound basis, right? I always love to say, "Here at Cisco, we can talk about Cisco forever. We're really good. Cisco talks to Cisco really well." When we want to tell that marketing story, how much does that speed of development, the speed of on-ramp, more than any of us can keep up with, how much does that play into your marketing strategy, yours and Carrie's, as you try to tell the big story around this particular Cisco Live 2025?
Yeah. If you're visiting the show floor, we have an incredible Cisco showcase this year. It's for the first time where we've put all the business units. Every time you used to hear Cisco talk about Webex and collaboration and security and the network and observability and, "Oh, we bought Splunk," right? It's all in one place. It's together. It's cohesive. It's a singular look and feel. That's the marketer in me. Particularly from a messaging side, it's about the use case. It's about what's in it for our customer. It's not shipping an org chart, and it's not talking about our company like it's an org chart. You don't need to know that, right?
You need to know what are we doing to help solve your mission-critical challenges as an IT leader today, as a practitioner all the way up through our CIOs and CISOs. I think you're seeing that. We're on to getting the feedback of, "What's it for the AI-ready data center? What's it for being able to future-proof the workplace?" Of course, digital resilience, all bounded by the AI conversation and, of course, that secure network we heard of today.
Fantastic. I also feel as people are entering the showcase, they stand there and they look around. Right in front of you, you've got the center stage, and you've got the great talks coming there. You've got your lightning talk theaters and your AI theater. Then you've got your demo, your demo, your demo, your NOC, and your SOC. It's all piece of a puzzle. I think a subconscious story that it tells is that your investment is safe. Wherever you happen to be, wherever you are on your data center journey, on your journey to the cloud, the investment you've made in Cisco is so wise and so intelligent because stand and turn in a circle. It is all available to you. It all communicates. It is all secure from the core.
Yeah. Being that trusted partner. I think Jeetu really hit that home this morning in the keynote. He really talked about fusing that security piece into the fabric of the network. I really felt that today. I hope you did too, right? That is something that he and that team have worked day and night on, right, to make sure it's in our hands of our customers. It is, it's protected and it's secure.
That's what makes it such an exciting time. I've got one final question before I let you go. Now, what do you have to do? You got to put on the hat. You got to start thinking down the line. We got to start thinking of what's going to happen in Melbourne, what's going to happen back in Amsterdam. How do we take this and continue to build this story? Because if we're talking about the speed of development today, imagine where it's going to be six months from now.
I know. I mean, Kevin Weil, the Chief Product Officer of OpenAI, I saw him speak at an event. He talked about at OpenAI, they do not have product roadmaps more than six months because it is so difficult to predict. I almost feel like we are in that space too. There is a, do not get me wrong, we have our roadmap, and our customers can find out by talking and being here and being with our practitioners. It is that we are just, it is forcing all of us to work differently and to work stronger together to make sure we are just staying two steps ahead. Again, having that security piece at the center. Now, Carrie did mention this morning we had the limited edition hoodies. They sold out in three minutes at the Cisco store.
I'm sure Carrie and Brian were thrilled with that.
Yeah, three minutes. I know scarcity was a topic this morning, but three minutes. I do believe we're going to try to order some more. For Melbourne and Amsterdam, we will make sure the limited edition hoodie piece and maybe their customized freeze location. I mean, we've got to come up with something that keeps them unique each time. They're a collector's item.
Did you get your hoodie?
I have not.
I didn't get mine either. All right. You're right on the same page. Billy, such a pleasure. Thank you for coming in to join us. It's really been a delight. Please come and do this again with us and again and again. We so appreciate it. All right?
Thank you so much.
All right. That was one treat. I don't know how we managed to find the time to do it in this packed Cisco Live calendar. Somehow, our own Zonafia Williams had an opportunity to actually sit down face-to-face with Chuck Robbins for a little one-on-one insight into the mind of our Chair and CEO here at the show. Let's listen in on that conversation.
Hello. I am super excited to be sitting here to have a conversation with the one and only Chuck Robbins. Chuck, how are you doing today?
I'm good, Z. How are you?
I am doing great. We're back here for Cisco 2025, 40th year anniversary. What's top of mind for you?
I mean, first of all, our customers and our partners are all here. The first thing is, let's make sure they have a great week. They're going to get tons of education. They're going to get a lot of time with our employees and our executive team. I think it's probably the most innovation we've ever announced at any Cisco Live that I've been going to. We're pretty excited about that as well.
That's awesome. We're all excited about it. AI is the top of all of our customers' minds.
I've heard that.
Last year, we talked about taking them through their AI journey. What's new for us at Cisco? What can our customers expect? How are we going to take them further in their AI journey in this next year?
We're going through that journey with them together. I think we're all learning as we go. I think the big shift is that customers are moving from chatbots and queries for information to this notion of agentic applications where agents will actually be performing tasks for us and communicating to each other. When you talk to customers about where they are with AI, there are three things that are sort of impediments that we have to get right together. First is they have to have modernized infrastructure. They got to have the right networking, the right compute because these agents are going to be communicating from perhaps the edge of the network back to the data center. The performance has got to be real-time. We got to have modernized infrastructure.
We have to ensure safety and security, which we're doing a lot of work on in that space as well. The third area that our customers are looking at is I need the skills. We're also doing a lot of education and trying to help as we learn too. We're trying to help educate our customers, employees, and just share the things that we know to help them be successful.
Exciting, exciting. The energy in this room is electrifying. They're all expecting something great. What's the one thing that they can take away from this event? One thing that they can expect to take away from this event?
I think they're going to leave here, hopefully, believing that we're innovating at a pace like we haven't in decades. Second is we should be a trusted partner for them to build out their AI strategy for what their infrastructure needs to look like. The third one is we hope they leave here with the belief that we can help them secure their AI and leverage AI for their own security in ways that they didn't know they could when they got here.
I love it. I love it. Chuck, thank you so much again. We're super excited for all that we're going to be doing here at Cisco Live 2025.
Thanks for all the work you're doing this week to make sure that we have a great experience for everybody.
It's my pleasure. It's my pleasure. I'm so, so excited to be here today. We're looking for so much, so much more.
It's going to be fun.
I cannot even believe this, Z. You managed to, I don't know how you did it, how you managed to grab Chuck Robbins and get some one-on-one time. Not that I'm jealous. Not that any of us here are jealous that you actually did that. That was pretty spectacular and great to be able to hear from him. My friends, it's kind of hard to believe we are done with the broadcast day. You've all been spectacular. I'm going to start right away by saying, okay, so our newbies, we have two and a half. So you were with us before, and now you're back with us. But Z and Michelle, bravo to the two of you.
Thank you.
You were thrown into the fire. Man, you have just eaten it up like it's been brilliant. It is so great to have you back with us, Lauren. We love your energy and your dynamism. Rob, my brother, whatever. What can I say that I haven't said for literally decades at this point of what it is? Here's what I want to do. I want to put a little cap on what it is that we're doing now by kind of talking through what you have found special about today. Not just like a technology highlight, but something where you think, this is why this event rings in the way that it does. Lauren, only because you're standing next to me looking spectacular here in the pink, I'm going to come to you first. What do you think really made today ring?
You said don't focus really on the technology, right? I will say, of course, we overuse this. It is really the people, truly. That is the magic. Being out there on the keynote floor and people coming up to me like, "Oh my gosh, this is Cisco Live. I haven't been here for three, four, five years." People are excited to come here in this place. That's huge.
We feel it, right? Billy was talking about that a moment ago as well. Z, I'll come back over to you because I know that is your thing, the way people engage and interact one another. There is something about getting everyone together in the room, isn't there?
Absolutely, Steve. You know what really tugged on my heartstrings was when that conversation with the Intel reps, the Dorin and Michael, I tell you, I felt the love. Because this is what we do is about our customers, those that we partner with to make what they do successful. To hear out of their mouths, I really felt like a proud parent and to be part of this amazing organization, to be able to be alongside Chuck. He is so personable. He engages with his people. He loves his people. We love him. I tell you, that was the best part, hearing those customers' stories and seeing how we are really impacting their lives and making a difference in what they are doing. I'm looking forward to more of those, more of those.
We're going to get a whole lot more of it tomorrow and on Thursday as well. Michelle, I'll come to you over on the marketing side because, again, I just had a chance to sit down with Billy Hackenson. He's just incredible working on Carrie's team. When you look at the marketing of all of this, how do we tell the story of what not just Cisco is as an organization, but the story of what Cisco Live is and what it actually means to people?
I'm so glad you came to me for that question. I wanted to talk about this. What resonated with me today was Jeetu's message on Cisco's structural advantage. There are three pillars that he really hit home that for me, I was immediately taking notes on. The first one is the platform advantage, how well we do as a one Cisco portfolio. Two is our own silicon, the programmability of that. That really stood out to me. Last is that AI is not an afterthought. This is something that we have to continue to basically use and build from the very beginning. That really stuck with me.
Absolutely. Couldn't agree more. What G2 was able to do today, what I love is when you hear the opening keynote, it's, yeah, it's a marketing pitch. I mean, we know what it is. We're here to learn about what's new. We want to hear the new announcements. There's the ballyhoo and the hoopla. I can't remember, Zazzle or whatever it is that they're calling it here at the show. Yeah, we have those moments of technology. Ultimately, it comes down to why Cisco and what it is that we do differently, why only Cisco can be that partner for you, get you where you need to go, and why your investment with Cisco is such a secure and safe investment over the long and the wide term. Rob, I always count on you so heavily every single year.
I learn mountains from you every time you and I get the opportunity. Yeah, you put the grandma glasses on or you put them away.
I don't need them. I forgot they were there.
That they were there and popping up on the head. You're always sort of my eyes and ears to the inside of how everything interconnects with everything else and supports everything else and makes the customer safe. Does that make sense when I say that? That their investment is safe when they come in with us.
Risk, I would always say. Yeah. Because I think risk is one of those things you never eliminate. You can only reduce and accommodate. You can transfer it. You can do different things with it. When it comes to AI, I think there was some talk about pushing executives to make certain decisions about AI and how we move forward. Honestly, every executive I've met at outside companies, and then combined with the messaging I'm hearing here, is that there's no need to push anybody. We're all feeling extremely compelled to figure this out as fast as we can. What really impresses me is the fact that the old story is always flying the plane while you're deconstructing.
Learning it, yeah, here in the air.
Boy, nothing really works better than to describe the feeling of what we're doing. Because we're all trying to make decisions about AI. It's obvious what the benefit is. It's obvious that it's nothing like anything we've ever dealt with before. Now it becomes, who can execute on the value in such a way that doesn't put themselves in danger for losing track of it, right? Because we're learning about it as we have to execute on it. To bring this back around to Cisco and what they're doing here, that's kind of what they're doing. It's just I'm like, "Oh, wow. We're laying out a vision for something that's still kind of hard to see." I mean, some parts of it are definitely there. From a market perspective, it's like, "Well, that takes some leadership." It makes sense to me, of course.
We connect data. Data is the gold in this situation. We're going to execute on that. That's going to put us in a good place.
I love it. I think it brings us back to kind of those three core words where we talk about agility. We talk about adaptation, right? We talk about being able to constantly revolve around whatever the demand happens to be to act in real time. These days, as things are coming at us so darn quickly, speed is, I mean, we do not know where we are going to be. Billy said it a few minutes ago. We can never plan more than six months out. We would have to just rewrite the story again and again and again. You all are so brilliant. I have had a wonderful time with all of you. I think you are incredible at what you do. I am so excited for our next two days together.
I'm going to go to one of these cameras and kind of wrap things up for everybody on the live stream. You guys just hang here with me. I'm going to go back over here. I'm going to go back over here. All right. We have reached the end of the broadcast day. We have had an incredible time with all of you throughout the day. On behalf of the five of us, I want to give a very special, very quick, very huge shout-out of genuine love to our incredible Cisco TV broadcast team back behind the scenes here in the studio out at Master Control, all across this entire event. The people behind the scenes are the experts who make the real magic happen for all of you on the stream and help us do what we love to do.
We're just so grateful to all of them and to all of you for tuning into the broadcast. Have a great night. We'll meet you right back here on the live stream tomorrow morning for an All-Star Wednesday opening keynote. Thanks so much for joining us. Bye-bye, all.