This is a 10-year, $60 billion transaction that we just entered into.
Qualcomm unveiling a new data center roadmap. This is not your father's Qualcomm anymore. What would you say to those doubting the execution here?
I think in the next three to four years the company could transform just on the robotics bet.
Wow. It's a very ambitious move. Much needed good news for Qualcomm here. Do you think Qualcomm can have a meaningful share of the chip content in these humanoids?
We're best positioned to be that player.
When would you project this next iPhone moment to happen?
And we are live from Maui at the Snapdragon Summit. I am joined here by Akash Palkhiwala. He is the CFO and, may I say, the COO of Qualcomm. It's really great to be here with you. Thanks for taking the time. And I guess my first question was, was it your idea to do this at Maui so everyone, all the media, likes your stuff even more, or what?
You know, it's funny. I fought the idea of doing it in Maui before I came to the first event. And since I started coming to the event, I understand the appeal. I understand just the atmosphere that it creates. It puts everyone at ease. You have people who are focused on the Qualcomm story, understanding the products and the technologies, and you get two or three days of time to convey those things to the audience. So I'm a big fan now.
Awesome. Well, you've actually been at Qualcomm since 2001. So you joined this company right after the dot-com boom and then the subsequent bust. This was the flip phone era. So much has happened since then. You became CFO in 2019. Then you got the COO title in 2024. You know, you went through it all. I mean, the rise of the iPhone, one of the biggest technological revolutions of our lifetimes. And now you're kind of stewarding Qualcomm into the next chapter. You put a dollar figure on it: $1.7 trillion of TAM here across so many different things for this company. Phones, PCs, cars, robotics, and of course the data center. So I guess what I would ask you is, how would you have described Qualcomm to an investor back in 2001, and how would you describe it to an investor in 2026?
Yeah, it's interesting. When we started off, Qualcomm to me has just been an incredible journey, very exciting, always technology first. We used to be the smartphone connectivity company, right? And it was all about phones and it was all about wireless connectivity, 3G, then going to 4G. And it was a different era, and Qualcomm was the most important company in my mind, and I think most people would agree, in terms of delivering chips that did 3G or 4G in smartphones. So that was Qualcomm then. Today I think of us as positioned to be the most diversified AI and processing chip company in the world. Right? So if you look at our targets for 2029, we'll have three big businesses: data center, smartphones, and auto IoT. And these three will roughly be equal in size. And I think having an asset, having an investment that positions you for that kind of diversification across the two big transformations that are coming through — one, of course, being agentic AI and the broader AI revolution; second, being robotics — and Qualcomm being at the center of both of those transformations, I think just creates a tremendous opportunity for us.
Absolutely. And this story that you're pushing, it's got investors pretty excited. You're putting out some aggressive targets here. So just to recap: full year 2025, $44 billion in revenue, roughly around $12 of EPS. And by fiscal 2029, we're talking about $40 billion coming from non-handset. That $44 billion today, you're a smartphone chip company primarily, and you're talking about that same amount of revenue coming from these brand-new business lines — not all brand new, but kind of growth-oriented business lines that haven't really been part of the story. The biggest thing I'd want to ask you is: yes, investors are somehow maybe misunderstanding this. They're not pricing in this growth story for Qualcomm just yet, a little bit. What would you say to those doubting the execution here? I mean, it is an extraordinary transformation. What do you think the naysayers would be wrong about?
I think the naysayers, I'd say, a lot of them are saying, 'Show me proof points,' and I understand that. And I think we've delivered a first set of proof points for data center. For a lot of edge devices, I think it's less so. We've delivered an incredible business in automotive, and we're not done yet. I think the amount of growth that we'll see going forward will dwarf actually what we've seen in the past in that business. And then these new devices that we're getting into — think of it as personal AI devices and PCs — that's just a whole new category of devices that'll come through, and we're positioned at the center of that as well. The second area is this broad concept of physical AI, which starts with autonomous driving but then goes into robotics, goes into industrial. And these are things that exist today, but they're all dumb endpoints; they have not leveraged AI, right?
And I think as those markets transform, as robotics goes from fixed form factor into portable, into these various industrial applications — whether it's distribution or transportation — or you go into humanoids eventually, the amount of technology that's going to be needed will be right in Qualcomm's wheelhouse. It will be all about high performance, low power, and we're positioned to deliver on that. So I think I'll say the naysayers on the edge side are limited. I think it's more about, hey, are you going to be able to grow, as you said, in data center? And we've given proof points now, right? We just talked about an Amazon deal, very excited to discuss that. This is a 10-year, $60 billion transaction that we just entered into. A portion of that is committed. I think it gives us a very, very increasingly optimistic shot at achieving the targets we've set out for next year, which is $5 billion, and in 2029, $15 billion. We'll probably do better than that, given how we're progressing in these new areas. And then finally, I'll say there's a lot of new technology that we're going to deliver on the data center side because that market is changing. I think what is required in data center has gone beyond the GPU, and being able to deliver high-performance, low-power computing that solves different problem statements beyond just processing strength — and we're positioned to do that. So just excited. I think we're okay with people saying 'show me,' because we're going to show that over the next several months, and I'm sure it'll get reflected in the stock.
Yeah. Well, I like stories like this. I don't want to look at companies where people are already pricing in the future. I want to look at companies where it's not happening, and I want to talk to people like you to figure out, you know, if that's actually going to happen, and have these conversations to figure out where you're taking this thing. This event, Snapdragon Summit, it's primarily for what the business that Qualcomm has today, which is handsets, PCs. You know, you have some exciting announcements across this new 5G mobile CPU that's the first ever running full AI agents on device, doing inference on device, running models on device. Very exciting. I think it's something that consumers will demand on the PC side as well. You know, Google Book and Gemini is kind of a native feature, hardware basically just becoming an agentic product, as opposed to an interface you operate for apps. My big question here is, do you think that there's any underestimation in the market that PCs, mobile, it's kind of like a mature market towards the end of the life? Do you think that there is meaningful upside for the consumer to actually upgrade their hardware with these new products?
Yeah, I'm going to maybe try to address your question in a slightly different way. Think about the consumer device ecosystem today. You have the phone, you have PC. We think there is going to be a third very personal AI device that's going to emerge, and think of it as a device that's on your person, a device that can see what you can see, it can hear what you can hear, and you can have an agentic AI conversation with it. So Meta Glasses is a great example of that. I think you're going to see a lot of other companies looking at watches, looking at camera with earbuds, earbuds with camera. There's a whole new category of devices, and everyone who's making devices in that area is using our chip, right? So you can debate with me if that's going to be a 100 billion unit market or a 100 million unit market or a billion unit market. But there's this new category of devices that's emerging within phones and PC. Those devices are transforming. I think we've ended up consuming a lot of content, playing games, of course, using voice calls. But the next big step is this agentic transformation that we're talking about at this conference for all of those devices, where rather than have everything run in the data center, what will happen in the future in our minds is there's going to be an agent on the device that is the place where all the understanding of the user sits, of the content that the user has, the location, what the camera sees, what the audio is hearing. And that context is critical to the AI request that you're going to deliver to your device, right? And so whether it's 'book a flight for me to Seattle,' the phone is the thing that's going to say where this person is right now, what the calendar is looking like, and what would be the best time on a certain day that this person could be on a flight. And for an agent to sit on the device and really make the device completely different — like you went from phones to smartphones, smartphones to AI phones. And that's a change that's coming, and I think that's going to make it very interesting for people like Qualcomm.
For sure. I mean, what makes me most excited about Qualcomm is I feel like what's completely given no credence by the market is how big the future of personal computing could be with a new device.
So you know, Cristiano's talked about how you guys are working with all the big tech companies, the hyperscalers, on how they dominate the next layer of AI past the model, like onto the actual hardware, right? Is it a pin? Is it a speaker? Is it smart glasses? Is it earbuds? Right. Qualcomm's going to be powering all of that. I'd be curious to know, when do you think this device arrives? Like, when would you project maybe this next iPhone moment to happen? Because when it does, I see Qualcomm being a huge beneficiary.
Yeah, I think over the next six to nine months there's a whole series of devices that are going to come out. It'll be from the frontier model companies. It'll be from the hyperscalers in the US. It'll be from the handset OEMs. It'll be the hyperscalers in China and the handset OEMs in China. And you're going to see all of these companies — really every one of them is looking at multiple devices, not just one, because we're not sure what the form factor that's going to win. And so Meta is going to have glasses, but they're going to have a watch that does some of the similar functionality. You're going to see a lot of people now do earbuds with cameras on them. There's this — Microsoft has this companion device to your PC that they have talked about for company badges. People are putting cameras and audio microphones on it as well. So there's just this tremendous amount of innovation that's happening in this area, and one of these devices will hit the consumer sweet spot, and then it takes off.
And it doesn't matter because you guys are in all of that.
If I were to ask you the question of which do you think is going to win, you probably don't even care because you're in everything, and the market will decide, right? The consumer will decide.
That's right. And it might be different for different consumers, and that's even better for us. I think having this multiplicity of devices that give you similar experiences, and the consumer decides how they're going to consume it, I think just increases the probability that something like that happens. And you're going to see these devices come out over the next several months. So excited and looking forward to it.
Yeah, that's going to be crazy. Especially with like OpenAI is working with Jony Ive. We might touch on wearables later, but let's just talk about the data center, right? This is the meat and potatoes of the story. I mean, we're talking zero dollars today, $15 billion in full year 2029. I mean, brand new fresh revenue. That's a meaningful share of what Qualcomm will generate in the future. There's a bunch of things going on. You guys got a full stack here. You're going to do custom silicon, so you're going to design AI chips for big tech companies like Amazon. You have a CPU, a server CPU, of course, extremely pertinent with the rise of agents and the CPU-to-GPU ratio. You bought Alphawave Semi for the optical and connectivity side of things, which is really interesting. And then we also have your own AI accelerator, which you actually have deployed with the sovereign customer in Saudi Arabia. So at Investor Day in June, we're talking about that $15 billion number. I guess for me, what I want to know is what really needs to go right to execute on that number by 2029. Like, for example, what happens in fiscal year 2028? Where are you with the business by then? And then I'd love to know what the share is going to be. Those are a lot of different products, all with different margin profiles. What do you really see carrying the AI line, whether it be custom silicon, your own chips, or something else?
Yeah. So I'm going to give you a longer answer to the question because I want to set the context of the opportunity, right? Here's what's required in data center going forward in our minds. Power is a problem statement in data center. So there's a requirement for high-performance, low-power computing engines. The second is the hyperscalers are looking for different suppliers and to be able to really kind of fragment the supplier base, and so that creates an opportunity for us. They're looking for partners who are good at technology and can deliver at scale, which puts Qualcomm in a sweet spot. Within AI accelerators, there is a need to deliver something that has very high memory bandwidth, because when you're doing inference workloads, you have prefill and decode. The decode workloads require very high memory bandwidth. Today, what exists in the market is really high compute engines and a high amount of memory, but you're building a two-lane road between two very massively populated buildings. And what is required is, is there a way to have a 10-lane freeway rather than a two-lane road between those two, because the workloads keep going back and forth. And so those are things that are happening independent of Qualcomm. But the reason why that's important is that it sets Qualcomm up to go and exploit the transformation that's happening in data center. So now let's come back to Qualcomm. As you said, there's four kind of separate product streams, and in each of these cases we already have proof points of success, right? In custom silicon, we've talked about Amazon, of course, and then we also have another global hyperscaler customer. So it's two customers that are really at scale now for us going forward, starting in 2027. Within CPU, we're going to take our 5 GHz CPU at the edge that is extremely power efficient, because we had to make it that for phones, and we're going to bring that to data center. We have Meta that has committed to using that CPU. It's a very large hyperscaler, obviously, creates volume for us, and now with their own very good AI strategy, it seems.
Yeah, that's right. That's right. It's, by the way, great to see Meta. The third is connectivity, and as you rightly pointed out, we bought Alphawave. They already had a product, but they did not have the scale that these hyperscalers needed. So now when you combine their product with our technology and scale, this has suddenly become a very interesting area. It's an area that's tremendously in demand. People are looking for new solutions. We're positioned to do that. And then finally, on AI accelerator, we're solving the problem that I outlined before, is building a 10-lane freeway between those two stacks. And the way we're doing it is really we're stacking one on top of the other. So we're going to have a logic die sitting with memory on top and drill holes through it and create the 10-lane freeway in the same building. You're not even going outside the building. And so that creates this tremendous differentiation and solves a very important problem for hyperscalers. So very optimistic on this stack of four different product areas that we're creating in data center. And the way I think about revenue ramping, to your last question, most of our 2027 revenue will be the two hyperscalers on custom chip plus connectivity chips. And so that'll drive most of it. We see almost like a linear growth between 2027, 2028, and 2029 to get to $15 billion. That's our target for 2029. And then when you look at the target for 2029, we'll have the two hyperscalers carrying forward, but then you have Meta CPU coming on top and AI accelerator coming on top, and that's really what makes up the full $15 billion. So very optimistic. I think the opportunity for us is much larger. We talked about having this ambition of 5% of a trillion-dollar TAM, which is obviously $50 billion. There's no reason why we're not going to get to that number.
Yeah. I mean, you mentioned that, gets me excited. You mentioned this other hyperscaler. We talk about hyperscaler. Are we talking about an American hyperscaler, or what's the difference?
So what we've said is we have one US and one China hyperscaler. So with Amazon being the US one, this would be the China hyperscaler that we're working with.
Interesting. And when you talk about that second hyperscaler, do you think it would be like a similar setup to the Amazon deal, where it's a custom silicon, or...
It's custom silicon. And in both of these cases, we're working on the primary accelerator for these players, right? So this is not an ancillary chip. It's really the main accelerator for those players that you would know about, and they're consuming that at tremendous, tremendous scale.
Wow. So you did touch on, you know, stacking that low-power memory next to the chip, you know, for trying to solve this insane memory bottleneck that we have. High-bandwidth compute, even today, was talking about scaling it to phones and cars and all that, which is tremendous. How deep of an architectural moat do you think that is? And I know Amazon was interested in that, so like...
I think it's actually a tremendous architectural moat, because what is required to build something like that is someone who understands the system, right? Because this is not about designing memory, or this is not about designing logic and putting them together. This is about creating an integrated packaged solution that cuts across the whole thing, that is designed as a full system, that has the ability to do repairs on memory to maximize yield, to maximize manufacturability of the solution. And one of the other things we're doing is to deliver this at very low power. We are bringing in LPDDR memory, low-power memory, rather than HBM, because those workloads don't require the performance of HBM, right? And this is within inference, it's decode workloads. So it's pretty unique. Now, of course, at some point people are going to be able to replicate it, but we don't think of this as like one-and-done. We have a long-term roadmap where the performance is going to continue to improve at a rapid pace, and very excited about bringing this innovation. And a little bit of what you've seen in terms of recognition of Cerebras as an alternative supplier, the acquisition of Groq that Nvidia did, is all trying to get to the similar problem statement. And we think our solution is by far the most elegant, and I think most hyperscalers would agree with that.
Very interesting. So the AI stuff is so exciting. I can't wait to watch your guys' progress there. But the business that's actually growing today is automotive, which is really crushing, actually. Up 61% last quarter, 23 straight quarters of double-digit growth. You raised your guidance on it. Qualcomm's in a lot of cars. I don't think a lot of people even know that Qualcomm's in their car. Again, this is a market that's kind of like phones and PCs, where it's not expected to, you know, 10x in the next five years. What is Qualcomm's story in the car, and how does it work? How should people think about it?
Yeah, so maybe the way you should think about this market is the part of the SAM that we play in within cars that is growing super fast, right? Because we're talking about the technology that's transforming cars. We're in three areas within cars. We have connectivity chips; most cars are already connected today, so that's not a big area of growth. The second is the cockpit experience inside the car. If you think about what cars used to look like from the inside before and what they look like now, it's completely transformed. And we haven't even seen agentic AI come into the car, because of all the places where you want to have a conversation and not touch buttons, cars are at the top of that list, because your hands are already busy.
