Welcome back to day three of The Six Five Summit: AI Unleashed, the final day. Over the past two days, we have built from the foundation up, and today we close it out. Day three brings it home. We are opening with Gary Dickerson from Applied Materials on how semiconductor innovation is accelerating the next wave of computing.
Then, across two tracks, cybersecurity and semiconductors, we are going to hear how the security lens and the silicon lens close the loop on everything we discussed this week. Daniel, we spent the whole week discussing AI apps, agents, platforms, but if the $800 billion CapEx bet is not going to pay off, it comes down to what happens at the wafer level. Gary Dickerson is closing us out from the most upstream seat in the industry. He sees everything. What is the silicon story no one in the AI conversation is talking about? Where does the real risk sit right now? Memory, packaging, capacity, or all three?
No doubt it is all of the above, and everyone can see that. Then there is more. It is energy, it is policy, it is regulatory, and of course, capital. We are going to hear all of this, and we are going to just get an insider's view, and it is great. It was such a fun time talking to Gary. I cannot wait for everybody to see this.
All right. Let us finish strong, Daniel. Here is Gary Dickerson. Welcome to Six Five Summit 2026 AI Unleashed. It is amazing the impact that AI is already having today, and I am even more enthused about what it is going to do in the future. There have been a lot of industry discussions, a lot of discussions around semiconductors, but less about how they actually get built, the equipment, the materials that go into manufacturing, because quite frankly, none of this would be possible without companies like Applied Materials. With that, I would love to introduce Gary Dickerson, President and CEO of Applied.
Well, thank you. Happy to be here with you guys.
Yeah, great to see you.
It's great to have you here. It's great to have you kick off day three for us here at The Six Five Summit. I think what you were trying to say in the beginning was these people that are creating little miracles every single day, right?
Yes.
That are changing our world.
Big miracles.
Yes. Many small ones and many big ones. We absolutely agree. I want to ask you something. Clearly, you are building EPIC Center right now, investing $5 billion here in the heart of Silicon Valley, to revolutionize the future of chip making. You feel very optimistic about AI and where we are heading with all this stuff.
Oh, absolutely.
Where are we with AI? This is a question I get asked a lot when I am doing interviews. I am flipping the script here. I am asking you the question.
Sure.
Are we in the first inning? Are we in the fifth inning, or is it halftime? Are we in the pre-game? What do you say when people ask you that question?
Yeah. I think AI is the biggest inflection of our lifetimes. I think it's going to transform every industry. It's going to touch every individual. For me, I look at how we're implementing AI inside Applied Materials. We're innovating the way we innovate. We're innovating the way we bring products to market. We're innovating in our operations and supply chain, in our services, across the entire company. The big focus for me is not headcount savings. It's bringing products and services to market, miracles to market, faster, and driving top-line growth. What is super clear to me is that the opportunity is tremendous. We are bringing products to market faster. We have a service business that's growing at a 20% compound annual growth rate.
I could bring services to market because these tools are incredibly complex. Our customers want to innovate with AI computing, with new materials and new architectures, but also they want a yield as fast as possible. It's really hard. AI really is a game changer in our ability to bring those services and products to market faster and better. I look at ourselves, where are we at? What inning are we in?
We are bringing products to market this year that we wouldn't have been able to bring to market without AI, and I think that this is just profound. I think that we will transform every part of the company, and I think this technology is at the foundation of competitiveness for everybody. It is the race for AI compute leadership is profound for every country, for every market, every company, and every individual. I believe we're very early in what's possible. Again, AI will have a profound impact across everything that we do.
Right
This race for AI compute leadership is the most important race that will determine who wins and loses.
I'm putting it in the pre-game.
Exactly. We are very early. I love that. I'm getting excited now, Gary. This is great.
Me too.
I got to tell you, though, everybody seems to have an opinion on the demand for semiconductors, and it is not just, oh, the next week, but it is the next month, it is the next year. You have a view that very few have in the ecosystem. What are you seeing in terms of demand, and where do you think it is going? Why is it different than the industry cycles? I mean, mainframe to mini to client server.
Yes.
Local and mobile, social, web, e-commerce.
Yes.
A lot of runs there.
Mobile boom bust.
Exactly.
I think this i s as you said, we've been through mainframe to PC, then you've got mobile social media. This is way more pervasive than anything that we've ever seen, and it's more profound. It really is at the foundation of competitiveness for every industry. The computing demand, and you also have, it's not compute just being consumed by people.
Yeah.
It's also compute being consumed by agents.
Right.
I profoundly believe that the demand is going to continue to increase as we keep driving forward the new AI compute architectures and the innovation, and there is more and more adoption of these technologies in every industry and in every company. I mentioned what we're doing inside Applied Materials. There is zero question in my mind, at the foundation of competitiveness. I'm very confident that this inflection will determine who wins and loses around the world and in every single market and every single industry. The compute demand is going to keep going up. There is zero question in my mind. From the data center to new edge AI types of applications, you're going to see more demand for compute, and this is inevitable because the economics are so compelling.
Every device, every country-
Absolutely
every industry, every workflow that currently exists.
Absolutely.
Yeah.
Subtly, I think he's telling us that that demand curve goes beyond weeks, months, years.
Oh, yeah.
It's into decades into the future.
Absolutely, and what we're seeing currently in our business, again, the race for compute leadership-
Right
is the most important race. In wafer fab equipment, around 80% of the growth this year is really in the areas that are being driven by AI. So you have advanced logic, memory, including High Bandwidth Memory, and advanced packaging, the multi-chip connectivity.
Yeah.
That's about 80% of the wafer fab equipment growth, and again, it's really being driven by AI.
Right
We're number one in every one of those different segments that are enabling that AI compute innovation. We're seeing tremendous pull from all of the companies that we work with, our customers, and other companies across the ecosystem.
Gary, it's not just a race to create more chips, right? It certainly feels like we need more of everything, right?
For sure.
Having said that, everything you just mentioned, breaking through memory walls,
Yes
building next generation advanced packaging to make everything more efficient or move data
Yes
via light. Of course, computer architectures that bring more transistors and
Yes
more throughput, all those different things. You see it all.
We do.
That's part of why you're doing what you're doing. That's part of why you're making the $5 billion investment in EPIC Center.
Yes.
That's part of why you have all these partnerships.
Yes
where is the opportunity that you see in all these different things? Where's the biggest opportunities lie for us to get that
Yeah
that efficiency? Because if the AI demand is as big as you and I, and by the way, Pat, we all think this way
Yes
is the compute, and the memory, and the networking, the storage, the energy, we're short everything.
Yes.
And we're going to be for the foreseeable future.
Yes. I think there's a real drive on tokens per second per watt. I've been in the semiconductor industry for decades. I love what we do. As you mentioned earlier, we're delivering miracles. I tell everybody your smart-
I think I stole that from you earlier, by the way just to give you-
Your smartphone, the application processor is a computer that has tens of billions of transistors, 60 mi of wiring, and it's over 1,000 steps to build, and it costs $30.
Yeah.
You think about what we've been able to create in this industry, it is a miracle. I can guarantee you, there are many more miracles that are coming, and by the way, they're coming faster than ever. I think what you see with a lot of companies is this concept of co-innovation through the whole technology stack. No one can drive this innovation by themselves. I think of it from materials to systems.
For us, we're enabling new innovations, miracles, in every part of that technology stack. You think about advanced logic to process data. There's tremendous innovations there. In some of these materials that we're creating, they're a nanometer or 2 nm thick, and you need to control to within an angstrom, which is a 10th of a nanometer, a millionth the size of a human hair, and we're just bringing all of this materials innovation together to enable that computing innovation.
In memory, you want faster and faster memory to get data on and off the chip, a lot of the innovation that Applied Materials enabled in logic chips are being brought into memory chips because, again, for AI, that high speed is incredibly important. Or stacking memory chips, where you're bringing the chips closer and closer together for higher speed- lower power consumption, and there's just so much innovation there.
Then you have advanced packaging, and that business for Applied is growing over 50% this year. How you connect computing components together is one of the most important innovations in the entire industry. If you look at an advanced package next year, you'll have over 300 chips in the package, you'll have over 500 billion transistors, over 2,000 mi of wiring, and how you connect those compute components together and how you move the data has an enormous impact on performance, power, and cost. For us, these co-innovation relationships are really fundamental, and I really think about this concept of high-velocity co-innovation. Again, winning compute leadership is everything to everyone in this ecosystem. Whoever's winning
Right
in compute will make an enormous amount of money. For us, we're working across this entire tech stack, in partnerships with our customers, with our customers' customers, enabling these new architectures in all of these different areas. When you say, "Which one is most important?" I would say they're all important because again, everyone wants to drive at the highest possible velocity in every part of this technology stack, and that's where EPIC, we made this $5 billion investment co-locating innovators together so that our innovators are, with our partner innovators, moving as fast as we possibly can, where we're creating new ways to build those chips. We're creating new ways to connect all those computing components. Our customers, again, the first place they're going to have access to those key innovations is right here in Silicon Valley.
I can tell you my experience being in the semiconductor industry for a long time, our leading customers, they want first access to the innovations. They want first access to the key innovators. At Applied, it's not just equipment. We have integration innovators.
Right.
We can work, and we are co-creating architectures with our partners that are dramatically better than anything that exists today. That high-velocity co-innovation here in Silicon Valley, where you have the highest innovation density per square meter of innovators, is a huge driver for Applied Materials and for our customers.
Gary, in EPIC, you have your customers, you have your customers' customers, you have your supply chain
We do.
all interacting here. Give us the future state. Looking at the vision of why you did this,
Yes
what do you want, three years, five years from now, what do you want to have accomplished with this?
It's really bringing these compute innovations to market faster. The economics are tremendous, especially AI. There's so much demand for compute. So improving tokens per second per watt
Right
cost per token is something that everybody is focused on, and I said earlier, my view is that AI demand is going to continue to increase, become more pervasive going forward. So every one of these companies, it's really time to market for those innovations and design wins. So for them, they're all in a race. Whether it's a logic company or a memory company, or how you connect all of this together, so you can have design wins in your business, or system companies or AI companies.
All of them, they're trying to drive this compute innovation as fast as possible. So for us, the vision is to accelerate those innovations dramatically instead of working in kind of a serial process as you go from one step from materials to systems to another, to another, to another. We want these innovators working earlier, faster and better. Here in EPIC Center, you have more materials innovations together Applied Materials and our partners than any place in the world.
This is the place you are going to have first access to those materials innovations that will enable those new architectures for compute. Our vision is we are accelerating time to market for those innovations. We are enabling our partners design wins in really big markets. Of course, for us, we want to be designed in with our equipment and services in this new platform.
It is interesting. It is like the world's largest technology incubator.
It absolutely is.
You remember when the VCs would stick all the companies in one building.
Yes.
You have done this in a very specific space, and with great scale.
Let me say one more thing. I think that there is the innovation, so we talk about innovating the way you innovate, which is really important. You also have to commercialize. That is the IC of EPIC Center. So our customers, to ramp any of these new architectures, they need to go to very high yields very quickly.
So it's not just who innovates first. Who is it that can ramp at high yield? When our customers are focused on design wins, they focus on performance, yield, reliability first. If you don't have those three things, cost doesn't matter. So whoever gets there first with those capabilities will win. And so we are working on innovations, but the innovations need to be able to be commercialized at high yield and the right cost. And it's really innovation in performance, innovation in yield, innovation in cost, all of those areas we're focused on with our partners.
Yeah, EPIC Center is as much a business proposition as it is-
Absolutely
an innovation proposition.
Absolutely.
The words, this is not a science project-
Oh, absolutely.
We're going. I would certainly hope not at $5 billion.
Oh, yeah.
Yeah.
No, we have very good line of sight to enabling the architectures, and then design wins with our products and services.
Right.
I would say the other thing that's really valuable for us is our ability to work with the architecture innovators and integration innovators. We have very high visibility on what's important over the next 10 years, over many technology nodes. It gives us very high visibility where to invest and where not to invest.
Right.
That kind of engagement, for us, is incredibly important. Obviously, enabling our customer success is important, all of those design wins. Design wins for Applied Materials are important, but I do believe it will make R&D way more efficient. If you think about how much money all of the ecosystem spends on these architectures, they are many, many, many billions of dollars. If you can bring those architectures to market faster
Right
and better, then the economics are overwhelming, not just for top-line growth, but it can make your whole company dramatically more efficient.
Well, Gary, really appreciate you taking all the time with us today. As our viewers head into day three here at The Six Five Summit, you are the opener. Big thoughts. What is the one big, I have heard a few messages here, by the way, throughout your conversation, but the one big takeaway on AI in the semiconductor industry that you want all of our viewers to come away with today?
I have been in the semiconductor industry delivering miracles for many years. I have never been more excited at any time in my career. I think the magnitude of the value that we are going to create is bigger than ever. The opportunities are tremendous. AI is the biggest inflection of our lifetimes, as I said. It will impact every country, every market, every company, touch every individual. The compute demand and the need for compute innovation is bigger than ever. I think the opportunity is there, but I believe that we have to innovate the way we innovate, or we will not accomplish what we could accomplish together.
Fortunately, w hat I'm seeing is tremendous pull from ecosystem partners in high-velocity co-innovation. We've built capabilities inside Applied Materials over many years that put us in a position to engage in a completely different way than we ever have in the past. I deeply believe I love material innovation, and I deeply believe material innovation will be the key to these new compute architectures for AI. It is so much fun. The opportunity is tremendous, and I look forward to working together across the whole ecosystem to create many big and small miracles.
Well, Gary, thank you so much for spending some time with us here. Appreciate you having us to your HQ and over to see EPIC Center. The first equipment may actually be being installed now
There we go.
as this conversation is going live. But love to continue to follow the work and, by the way, love the enthusiasm because
Well, wonderful.
we feel very much the same way here.
Wow, that's great.
Yeah.
Thank you.