Hello, everybody. Oops, sorry. Tim Long here, Barclays, IT hardware, comm equipment analyst. Thank you for joining us today for Motorola Solutions presentation. Happy to have Mahesh here, who is EVP CTO of the company. He's, you know, came over through Avigilon, so certainly one of the more exciting areas of the company. As you know, hopefully we're big, big fans of Motorola and particularly what's going on kinda under this umbrella of software and video and AI. So let's get to it. Maybe Mahesh, if you can just, you know, at a high level, you know, Motorola traditionally, you know, viewed as a radio voice company, but has really grown massively in these new areas. So maybe just talk a little bit about how you're helping build that.
I know you're in Boston, so there's more, you know, more, you know, solid candidates to bring into the company. So just start at a high level and then we'll, we'll get.
Yeah.
Into some of the specifics.
Sure, sure, and by the way, just to, in case any of my Chicago teams or anybody else is watching, we got great talent everywhere.
Yep.
It definitely helps the cause. So, as Tim explained, I came into Motorola in 2018 through the acquisition of Avigilon. And so I had the opportunity to come in with the benefit of a fairly significant video team. And one of Avigilon's claims to fame was really computer vision and analytics, really both built into the camera, but also running server side. And along with that came the people, the talent, a fairly extensive AI team that was able to take a lot of what we did with video and also make it more broadly applicable. Prior to the acquisition of Avigilon, Motorola had already taken a step towards command center software, everything from our acquisition of VESTA to Spillman and such.
and what became interesting with the acquisition of Videotec into this fold here and then subsequently the series of acquisitions we have made, is really bridging this tenuous gap between enterprise security and public safety. If you think about it, when events happen, on the premises of a Fortune 500 company, typically the way public safety gets involved is through a 911 call.
By creating automations with video, early detection capabilities, and then bridging that into what public safety can effectively respond through 911, through dispatch solutions, through records, through law enforcement coming in, and making sure that it's not just a single lane of information traffic, which is that phone call, but really everything that we can enable today with video directly going into PSAPs, Rave allowing every individual there to have an application where they can push a panic button, where they can interact more effectively with PSAPs, for PSAPs to be able to talk to each other in that time of crisis.
What we're able to do is really make sure that that 60 seconds saved to save 10,000 lives per year through an effective 911 response benefits from the fact that we get that early detection. We get that better information exchange from public safety to enterprise security.
Right. Great, great. So maybe let's start on AI. Can you talk a little bit about how Motorola, you know, is using AI in its products and solutions?
Sure. And for us, this may be a bit of a crude analogy here, but Tim here's drinking coffee. It's black.
Yep.
It almost looks black, like black coffee. Historically, AI is like the artificial sweetener that you put into coffee, right? You may use that instead of regular sugar. Now, the artificial sweetener comes with certain benefits, right? Zero calories, but there's also some disadvantages to it. Some say that artificial sweeteners are a carcinogen and that maybe it's slightly bad for your gut health. Historically, AI has been that artificial sweetener. It comes with certain very clear benefits, and it also comes with certain disadvantages as well. Really, over the past few years, generative AI has sort of changed the equation a little bit where we're talking less about AI being the sweetener, but actually the coffee itself.
And so it is through that lens that I approach AI within Motorola Solutions, where it's not an ingredient per se, but it is something that is very much focused on enabling very clear outcomes. We have cameras, we have radios, we have command center software, we have video software. Everything depends upon having this large pool of data that can be brought to the surface in that moment of need. And AI really enables that to happen, and that's also one of the fundamental outcomes we want to enable with our software.
Over 90% of our cameras today are AI-enabled, in some form or the other, all with the hopes of saying, "Hey, video is not just a retrospective investigative tool, but video is actually that, the camera has the capacity to alert you, tap you on the shoulders, and say, 'Let's look here.'" In the enterprise security world, when the majority of our customers actually do not have a security operations center, they don't have dedicated people staring at a video wall. When you have people staring at a video wall, after about 20 minutes, there's about a 20% chance that you will see something important happen. That's just after 20 minutes. Imagine after an hour, it actually drops down to about 2% or 3%. AI, in that case, actually helps you focus attention effectively.
But when you do it well, you don't need that one dedicated person sitting in front of a workstation looking at it. Mobile apps become a big deal in that case where, in a school, it is typically a school administrator who receives an alert that, "Hey, there's someone walking in with a weapon potentially," and that's, and there's an action that is required. That person's primary job is not security. That person's primary job is something else. So we democratize how effective security can apply regardless of whether that enterprise has the capacity to have a security operations center with dedicated people staring at video. Now that extends to, say, 911. We launched our VESTA NXT solution quite recently.
In that world, and this is an I'm combining two examples together for this anecdote, but a person comes into a large city. That person has a child with special needs. Momentarily is distracted, that child walks away. Now that child is lost in the city. There's a 911 call that gets made, but that person actually does not speak English. That person happens to speak Portuguese. Real-time translation with 911 enables that Portuguese to English translation to happen in real time, and by the way, English to Portuguese so that person understands. But it's not just a translation aspect of it. It is understanding what is said. So now very quickly we can say there's a description of that child that's being given in this conversation.
That description can then be pushed to our cameras to say, "Do any of these cameras actually see someone matching that description?" and when that match is seen, then appropriately, whoever is the resource, the responder resources that are close by, perhaps carrying one of our radios, perhaps with one of our applications on their phones, gets an alert that says, "Hey, this person matches this description." By the way, through the course of that conversation, there's also something that says, "Hey, this child has special needs." So if that child has special needs, then we also want to be able to let that responder know how to properly approach that child. And all of that context extraction and activating the data that's associated with this was a proper way to approach the child.
What is the right protocol for doing this? It is not something that people are actively pushing to that responder. It is automation, AI-enabled automation that's pushing that out. And if you flow that through the benefit of data, over the past 10 years, the mantra in public safety has been de-silo data. And I always like to say that previously you were blind because you didn't have enough data. Now you're kind of blind because you have too much data. And what AI and generative AI is really doing for us is activating that data so it's actually something that is helpful in that moment of need where the context allows us to separate the relevant from the irrelevant. And that is how we are taking advantage of data.
All of this, from a clear picture standpoint, because we attack every single element of the incident life cycle, we have a 360-degree view of that incident. We have the data that sits behind the 360-degree view of incident. AI then for every user, in that incident life cycle, whether it's someone in a PSAP, someone responding, or someone in the enterprise, we have the AI capabilities to surface what is important and help automate simple workflows in the background so that 60 seconds saved becomes a lot more than 60 seconds.
Right. Yeah, that's a great overview. Maybe, you know, you're not a finance guy, but talk a little bit about the differentiation of offering this across, you know, the products and services that you have. Obviously, ability to maybe take market share because you have better technology or better ecosystem. How does that play into, you know, the kinda core strengths of Motorola, and having, you know, as you said, this 360-degree view? How does that manifest itself in kinda the business end of it?
So, AI tends to be hype if it's not supported with the right sort of contextual data that sits behind it. That is why this whole 360-degree incident view is so critical to everything that we do, because now we can actually put AI to work, to help provide, produce the right outcomes. Think about it the following way. An enterprise that bought a bunch of really smart cameras, maybe hundreds of these cameras, there's an integrator who sets up these cameras, and they have to anticipate what are the types of events am I really interested in detecting in each of these cameras. That takes time, like minutes, perhaps even hours, per camera to go set up. If you have 1,000 of these cameras, it becomes really cost-prohibitive for someone to actually leverage AI and set up the advanced alerts.
In the end, if you have the right bridge going into public safety and through public safety's records, information, and others, you have an idea of, "Hey, what are the types of events we're seeing in this geography?" Maybe there's incidents of stolen vehicles. Maybe there's incidents of vandalism. Maybe there's an active shooter that was seen somewhere in this geography. All of those dynamically can inform what the camera should actually be looking for in that given moment. It, by changing what used to be a task of anticipation to a task of proactive analysis and reaction, we can now make every one of those AI-enabled cameras without those hours of setup required ahead of time to actually now bring forward a capacity to better protect that facility. That is a critical differentiation for us.
In the public safety side, as we are thinking about everyone going everything going from 911 all the way up to case closure, that incident view makes a very big difference. When that officer is responding to an incident and they have an APX NEXT device in front of them, we just launched SmartIncident, which is an application for the APX NEXT apps, directly connects with our CAD solution, CAD integrated with our 911 solution. In a single screen, we wanna be able to offer a summary of that event. That summary generated through a transcription of a conversation that happened with the 911 call taker.
that once, one-screen summary given in the context of that particular officer responding to that particular incident, so the AI is conscious of the fact that it is offering a summary to a first responder, a police officer in this case. By the way, that summary would look very different. It was an EMT. That summary would look very different if that person was a firefighter. The ability now to have the contextual information and the screen, in this case, the radio, which the majority of our first responders use, to effectively present that information in a synthesized way now helps them cut down the 60 seconds necessary to respond. That is a differentiator for us.
Right.
On that front as well.
Okay. Great, great. Maybe let's get into video a little bit. Obviously, you came from Avigilon, so.
Yep.
Near and dear to your heart. There's been a lot of other M&A that's.
Yep.
Been done, you know, putting this more complete solution together. Talk a little bit about where we are, maybe first from a hardware standpoint and like the different markets that you're hitting. Then after that, I do wanna get into about a third of that business is software, so I really wanna get into the, you know, the video analytics software piece of the video side.
Yeah. No, so I would say even when you think about the cameras, the intellectual property of those, in that camera is actually software.
Mm-hmm.
Right? It's every camera that we hang off of a facade of a building or anything that you see here in this hotel. Those cameras, if they were ours, are actually mini computers of various sorts, capable of running AI, capable of running multiple key applications right on board. What we have done since the Avigilon is make a series of acquisitions. IndigoVision helped us expand our international reach, and has now the solutions folded into the main Avigilon portfolio. We acquired Alta, or sorry, Openpath and Ava, which together became Alta, which were the cloud analogs of our solution. And the hardware, by the way, Avigilon came with the broadest family of AI-enabled cameras in the industry. Alta Video, Ava, came with one of the best user experiences for cloud software.
Now we're attaching the portfolio of cameras that we have within the Avigilon side to our Ava to Alta Video, effectively now making sure that we have the broadest portfolio of cloud-connected cameras in the industry. Similarly, we ended up acquiring Videotec and Silent Sentinel, adding it to the Pelco portfolio, another one of our acquisitions. A good chunk of the market today is cameras, as you said, but also cameras attached to other third-party software. Certain verticals prefer that, certain specific types of software. We're approaching that first by saying, "Hey, if you want an explosion-proof housing on your camera, great. We have the best-in-class explosion-proof housing." That explosion-proof housing now contains within it our intellectual property when it comes to AI and other capabilities that are built into it.
Best-in-class thermal cameras, best-in-class solutions that work in hazardous environments of various sorts, oil and gas, and such being an excellent example. Silent Sentinel brings in the capacity for us to have wireless cameras that are installed in borders and such where there's vast distances between infrastructure. So a lot of them tend to be solar-powered. A lot of them tend to be powered with high-power zoom lenses in a way where you can cover great distances with single cameras. We're able to now go in and augment our hardware portfolio with these key sensor modalities in place in a manner where we can attack these significant high-growth verticals with unique solutions.
These unique solutions then allow us to add the rest of our portfolio into the mix from a hardware standpoint as well, all now also powered by this notion of assistive AI that sits in the background, which benefits from making sure we can apply the right context to analyze this data appropriately.
Okay. Great. Yeah, you did mention, you know, some of the software applications. I think people think video. They think a camera, not necessarily software and.
Yeah.
And applications. Maybe touch on some of the key, you know, use cases, applications for the software business. Obviously, license plate recognition is pretty large, anomaly detection. Maybe walk us through what are the offerings that you can offer to your customer who has a camera network, to kinda upsell some of these software solutions?
There are really three parts to it. There is everything that falls under the detection column. The detection column includes everything that we can do in terms of event analysis, understanding when somebody's loitering, someone's carrying a gun, a variety of threats of various sorts. ALPR is a very significant portion of that story for us because license plates tend to be one of those key pieces of information, whether it's a stolen vehicle, there's an Amber Alert, there is an incident of someone who is perhaps suspended from a school coming back into that school, a parent who is not allowed to be on property, a sex offender of some sort. All of this gets tied to a vehicle of some sort that tends to be the trigger for an alert. So ALPR is a big part of that story. You mentioned anomaly detection.
As I mentioned previously, setting up a camera to alert on the right two types of activity, it tends to be a very hard thing for folks to do. But anomalies, things that happen rarely, warrant some level of attention on the part of a human user. So that tends to be, that anomaly detection is one of the most powerful activators of the use of AI in cameras among our customers. So that's a very significant part of what we do as well. And then the second column would be analysis and forensics. This is either during the course of an incident happening that's in progress or post-facto or retroactively.
You want to make sure you can leverage all the video data that was collected, perhaps audio data and other types of data that comes into the fray as well, to be able to say, "Hey, where did this person go across my entire camera network as this incident was happening? Where's this person now?" The special needs child that I mentioned before. These are real-time analysis tools that, by the way, for the cost of what you spent already deploying your cameras, you get that capability as part of this equation, and that becomes a powerful differentiator where now all that data is actually useful, actionable in that moment of need, and so that search and forensics capability becomes quite important.
For public safety use cases, that ALPR data, being able to say, "There was a active shooter situation where this person showed up in three of these locations at these three time instances. And by the way, I don't know who this person is. I need to figure out who this person is as soon as possible." Understanding the vehicle that was present in those three locations, at those three times very quickly narrows this down to one or two possibilities, which then accelerates the law enforcement exercise of saying, "Who is this person?" and hopefully preventing something worse from happening. So that is how we monetize software in that case. And that's why our software applied to both our hardware, you very uniquely, but also with third parties. By the way, we have AI appliances, servers, and cloud analytics capabilities.
We acquired a company called Calipsa, which allows us to bring AI capabilities in the cloud to third-party cameras as well. This allows us to basically take dumb cameras that exist in a brownfield situation and pull them into our ecosystem as something that is actionable by our software.
Okay. Great. Maybe if we touch on command center software a little bit. You mentioned it throughout the talk here a little, but curious about the strategy there. You have some of the big modules. You're kind of putting them in one system. How did that and how is that integration? What are customers saying? And maybe on this talk a little bit about on-prem and cloud where a little more activity recently in customers looking to have more cloud-based solutions?
Yep. Yep. So, by the way, as we have been chatting here, I've very intentionally been blending some of the conversations between our critical communications portfolio, radios, video, and command center because as much as technology-wise, there, there are distinct applications, the value is in how we connect those pieces together. So as we think about command center, we have over 60% of the PSAPs out there leveraging our 911 solution. And 911 is one of those that key bridges between public safety and enterprise security. And we are, as I mentioned previously, really adding lanes to that data bridge between the two. So Rave was a key element of that story for us, both from the standpoint of allowing communities to be able to very effectively connect with public safety, but also in very targeted applications like panic buttons.
and communities in this case with Rave Alert, mass notification capabilities, two-way mass notification's probably a little bit of a misnomer because notification implies it's one way, but it's actually not. It's a two-way communication of information. And Rave has multiple statewide deployments today. That statewide deployments get inherited by schools, as well. Schools then deploy specialized capabilities like panic button capabilities for their students, for their teachers, etc. All of this connects effectively with our 911 solution, but also connects with our CAD solution.
What we are seeing as a trend, a very key trend going forward is that as more capabilities, the officers in the field, the responders in the field used to carry fairly simple devices, radios, and they sometimes had a mobile data terminal in their vehicle, and it was, in some cases, a clunky computer that did a whole bunch of stuff. They're able to do a lot more just using handheld devices today. So a lot of functionality that historically was in the purview of a PSAP is starting to migrate down to the edge. SmartIncident is a key example of what we did with APX NEXT, but there's more to that. And the Responder app is a key APX mobile application for iOS devices and also Android, which some of our larger police departments are using now as well.
That data going to the edge means that certain functions in the PSAP are actually starting to blend together. 911 and CAD is an example. So dispatch and call taking are starting to come together in some fairly unique ways. And as those elements come together, both from an on-prem standpoint, but sometimes also leveraging cloud, and hybrid has been a key strategy for us. And so we have gone into our customers and said, "Hey, you may have invested, and you may be happy with your current on-premises solution that we provide. We will be able to add incremental features to it, but those incremental features delivered via the cloud." And that's been a very successful strategy for us. The hybrid attach rate has grown quite significantly.
We are seeing some customers also now gradually migrate most of their capabilities to the cloud in that way as well. But that is coming on the heels and perhaps it's greatly benefited by this drift to the edge with more cap use cases going to the field because it simplifies the solution within 911 and within dispatch, the point where in the cloud we can bring those unique capabilities in very simple ways together. Records, evidence, etc. now become part of this data platform that is uniquely activated by AI connected to this real-time use cases for 911 and CAD as well. With command center, it is all about this notion of the real-time response for us because of our overwhelming presence in PSAPs. That again I keep going back to 60 seconds saved saves the 10,000 lives.
That is the thing that motivates every one of our engineers out there because it's if I save a second, it materially helps us save a life. It materially helps us solve for safer, and with that particular focus in mind, we can figure out exactly how the data portion, records, evidence, etc., now makes that even more of a powerful story connected directly with the community with Rave, Responder, APX NEXT through SmartIncident, other apps as well. That is the command center story for us. But again, the boundaries between command center and video, the boundaries between command center and critical communications are radio. Those boundaries are very intentionally blurring because of how we're bringing these pieces together.
Okay. Yeah, you worked in the Solving for Safer. Greg would be happy that you got the new, new slogan plugged in there. We have two minutes left.
Okay.
Maybe just closing out on command center. Just talk about the suite of products, the ability to take, you know, the strength that you have in 911 and use the common software and user interface to do better in CAD or records or any of the other modules.
Correct. So, the key portions of the key pieces, I would say, of command center software, at least the way the industry looks at it today is 911, CAD, computer-aided dispatch, records, and evidence. Those are the three, four, most significant pieces of the story. The trend we are seeing is that from a user experience standpoint, 911 and CAD are starting to come together. It's more of a unified solution going forward. We're also seeing that, well, the dispatcher tends to interact with the record system frequently, and as the dispatcher's interacting frequently with the record system, there are elements of what the record system does that becomes naturally part of the user interface of a CAD solution, thus also part of the call taking solution if that's an integrated.
Yeah.
Piece of it. That's now flowing down to the first responder who has a body-worn camera. That feeds into the evidence framework as well. But that's also coming into that screen that officer has in front of them. So now that records platform is feeding the dispatcher. It's also feeding that frontline responder effectively as well. But that person's not only a consumer of information, they're a generator of information into the record system as well. So we're able to close the loop more completely with our customers with those four key solutions benefited from the fact that the pathways we have for the frontline responder to add to that data corpus and add to the workflow that the PSAPs are using within the command center, that becomes that much more effective.
Okay. Great. We are right at time. So thank you, everyone, for joining.
Thank you.
Thank you so much. Really appreciate it.
Thank you.
Thank you.
Appreciate it.