Qualcomm Inc (QCOM) November 12, 2026 Conference Transcript

Well, good afternoon, everyone. It's great to have you join us. I am particularly honored to be hosting Cristiano Oman, President and CEO of Qualcomm. Welcome. It's great to have you here. Thank you. Very happy to be here. We've got a number of things to hit on in a brief amount of time, and so I'll start with the first one, which is really the evolution that is underway at Qualcomm. You've been CEO for five years. You've focused the company on diversification beyond smartphones to automotive and IoT, and now doubling down on the data center. Can you talk about your broader vision for Qualcomm and what you think Qualcomm would look like exiting the decade?

Yeah, no, thank you. I look happy to be here. And so Qualcomm is a company in transformation and hopefully it feels a little bit different than, you know, the Qualcomm, which was the most focused company in mobile. I think the vision for Qualcomm is we, I think we have one unique capability, which is the ability to have both communications and computing technologies, and I think our DNA is high compute density, low power consumption. I think we have built a company thinking about battery-powered devices. If I have to summarize, the elevator pitch of Qualcomm is everything, you know, communications, computing, you know, for low power consumption. And we have a look of the surface area and how we're going to add new capabilities to the company. I think the first we look, it was a natural evolution for Qualcomm to diversify into the edge. I think we're very happy with the business we're building. It's still building, even though it's getting scale on automotive. I think we set ourselves well for the next transition that is going to happen across different industry verticals. I'll come back on the mobile in a second. And we have a right to play, and anything is coming from our success in automotive and robotics. Unlike data center, we're investing early in robotics. And data center will be a natural thing, I think, for Qualcomm when you think about scaling our compute. We believe that we're still in the early days of AI, and I think that's how I think about Qualcomm. Having said that, there is an interesting transition We have been very bullish on what's going to happen to glasses and new types of devices, including the phone itself, and I think that the next cycle will be coming. It's just very natural, and Qualcomm will participate in that as well. So that's how I think about Qualcomm right now. I'm happy when you ask me about the vision of the company, all of the uncertainty that has been, you know, I remember the conversation was about, yes, I know you have an interest in diversify, but let's talk about what's happened on mobile, Apple, leaving the business, the licensing business. I think we got ourselves in a position that we created some stability in our licensing program. We have gone to the Apple transition, and now we look at all the different vectors of business that are growing. We're excited about that.

So let's hit on the data center. So, you know, back in June, you outlined an aggressive plan, including CPUs, accelerators, custom memory, and interconnect, all of which are expected to be ramping by 2029. With this as a backdrop, you know, what does success look like for Qualcomm in 2029? I know you've talked about that $15 billion target, but maybe you can unpack that a little bit.

Yeah, no, happy to. Look, we took an approach on the data center, maybe a little bit of the premises. It is a very large stem. We will have discussions every year about the rate of growth, but it doesn't matter. I think it will continue to grow. It's a very large stem. I think we're still in the early days of AI. We saw how the data centers is aggregating. You're going to have a specialized compute to do different things given the scale. And we decided that we needed to invest in data center to be a player in the long term, not an opportunistic play. So we actually, I will say that we're being very conservative, and I'll be very careful. I'm going to choose my words because we're not making any assumption. We're not leaning forward. I'm being super focused on demonstrating to the investors that we're going to deliver the $5 billion in fiscal 27, $15 billion in fiscal 29, but we're investing in a roadmap for 50. We're just focused on 15. So we're investing on CPUs. We think the CPU will continue to be relevant. It will be very relevant. I think we're in a good position. We have a very good CPU IP. Our CPU is very high performance, low power. We have one customer, and that's in the 15 billion that is meta. But as soon as we have, I have a number of companies that said, as soon as you have silicon, we would like to evaluate. I think you're going to see a scenario that CPU and TCO will matter. We have a right to play there. We're building on the connectivity roadmap of Alpha Wave. We invested, I think Qualcomm has a communications, I think, background. We think there's a lot of opportunities, especially when you think about photonics, co-package optics, scale-up, scale-out. I actually like the Marvell acquisition of Lomintum. My analogy to that is what happened with RF front-end. When we acquire RF front-end, you start to see combinations. We had an RF front-end attached to our baseband. Some of the other players start working with other baseband providers. There's, I think, the laser plus, you know, silicon. Partnerships create an entry door, I think, for Qualcomm. We are going to leverage our scale for custom ASIC. I think it's becoming, you know, with the Amazon agreement that we get, and we have two agreements now in place, we can play in this and add our connectivity, and we made a bet on the accelerator. We made a bet on inference, dedicated SKU for both decode and prefuel. I think with our HBC, we're the very first company to do stacked DRAM with LogicDye, and I think we built probably the best solution for decode. Again, we have to have silicon. Everything changes when you have a real product. HBC is performing very well. I'm doing better than what we said on Investor Day. We said 0.5 picogials of energy per bit. I'm doing 0.3 in the lab. Our yield is already in excess of 52%. So we're just going to build it and excited about the opportunities. If I step back, very high confident number 5 and 15, we're just going to invest in a roadmap, and we're going to be a player.

Excellent. Now, I know you have a very broad portfolio, so it's hard to isolate, you know, one technology within all the different kind of markets you're attacking inside the data server, but I'm curious if there's one commonality. Is it, you know, power efficiency, inference cost per token, your ability to spin and re-spend silicon incredibly fast? You know, what would you highlight from a technological perspective?

Look, if I have to choose one, right, I think the first thing for Qualcomm is it's power-efficient computing, and it's interesting. I had faced incredible amount of skepticism when we said we're going to go do automotive. I even heard from investors, your NXP acquisition didn't go through. You're not going to be able to be successful in automotive, and automotive was perfect for Qualcomm, right? You're not going to put a server power consumption in terminals within the trunk of a car. You need connectivity, you need a lot of computing, you need low power consumption, and you need a company that can have capabilities across multiple technologies with high-level integration. That's gonna be even more true, when you think about robotics. You know, our silicon is running a lot of, in the automotive business, a lot of the assisted driving and autonomy stacks. That's inference, that's kind of edge AI. That's even more true in a robot. The connectivity, the power consumption, the compute density. So when I look at a data center, I'll say, Okay, we are going to rely on power consumption and its impact on TCO, thermals and inspect on TCO. The next thing is, we do a lot of chips. I think we have the ability to do a lot of chips, especially on the leading node. We have the order about, if I look at total number of tape outs that I have in a year, It's an inaccessible number between 60 to 70, so the ability, as you said, to quickly adapt, spin, and do bespoke solutions, it's one of the reasons I think we came in very late and we landed an Amazon custom ASIC agreement, you know, with multiple generation. I think that's another asset of Qualcomm.

So how about beyond 2029? I know you put the 50 billion, but you don't want anyone putting that anywhere near the model. But outside of that target, you know, what else should we be thinking about in terms of your longer term ambitions?

Look, we are very, very focused on 40 billion of known handsets for fiscal 29, and I think we outline, I think we see a path in the company, you know, within, as you go to the end of the decade, in initial years, to be at $100 billion, I think, revenue. As I said, I think we're positioned the company now for each and every one of our business that we're building to grow. If you look, all of them are positioned now in terms that are expanding that we have a right to play. Data center, we're going to have to prove ourselves. But I think that's kind of the position for Qualcomm. And I will not completely discount mobile because, you know, mobile is going to have an important role to play. We're starting to see the beginning of a transition in mobile. I'll say about mobile, that's not to spend a lot of time on handsets, but there's a huge cemetery of companies that have been in mobile. We never went to that cemetery, but you see many, many companies that went from hero to zero. The mobile market will change once in a decade, and I think we started to see the beginning I think that's the beginning of the change with AI and agents, with workload shifting to other devices, so that could create opportunities for Qualcomm. I look at that as optionality for Qualcomm, even though that's our core business.

So maybe a last question on the data center. As you gain greater traction with a greater breadth of customers, are we going to see a point in time where you have a gigawatt scale kind of project? And as part of that, would you consider financing like we've seen from some of your chip peers?

Yeah, it's very interesting. I think one of the most interesting questions these days, everybody's trying to find out, is there a Kelly Blue Book for price of chips because of what's happening, I think, with finance? So let me answer a question in a different way. I think we're investing so we can have gigawatt scale, you know, customers. I think that's what we're doing. We're not taking a opportunistic view on data center, so we're investing for that. I think we're open to a lot of possibilities. We're also very careful. I think we have to really focus on proving ourselves with customers that will help us get scale, and I'll probably leave it at that.

Sounds good. So maybe we'll move to your SpaceX deal announcement. I was down at Starbase in April where Elon discussed aspirations beyond Starlink and clear indication of competing in the cellular arena, and you are, in my view, the natural partner. Can you share more about the agreement and how big the deal could look like over the next three to five years?

Yeah, absolutely. Look, I'll tell a little bit of a story. I'm going to try to be short here. But this is very interesting. I actually go back to, I go back to, I think, 2020. 2020, it's a little bit, I remember because the conversation happened in the word economic forum in Davos, right before the news broke about the pandemic, right? So 2020, the US government came to Qualcomm and said, there's gonna be this transition in the seller industry eventually to a new architecture, And we like to have, you know, American companies investing in technology of, you know, infrastructure, you know, the cellular infrastructure. And especially this, we invented this business, and kind of we don't have the technology. So look, we believe there's going to be a transition to a different kind of architecture. Eventually, the base station is going to be built more like the way the Internet's built, the cloud is built, the data center is built now with AI. So we're going to leverage our assets, and we haven't talked much about this, and we're going to invest on silicon that could enable a very advanced radio unit, an accelerator card that actually inference to basically, you know, accelerate the digital part of the base station, and we put ourselves on the road to develop that. We didn't talk about it. It wasn't part of any of our main initiatives. We thought it was strategic bet. It was very interesting as we got ready with the technology, it coincided that, I think, SpaceS has this vision about building a modern architecture of cellular, I think, what would be an AI native? And I think that's where you started to see, I think, the forming of this partnership. So what this deal is, I think, SpaceX is building a nationwide terrestrial network. It's gonna be probably one of the largest deployments of terrestrial networks in the United States. We're providing the silicon that does all of the massive MIMO for 64 transmit, 64 receive. We're providing the accelerator cards. We're providing the layer one softer. SpaceX. It's designing, I think, several elements of the big station, provide the other layers of software. This is exciting. I'll answer your question, but then I'll talk about a bigger picture. To answer your question, it starts to show up in revenue for us at the end of fiscal 27. It's more of a fiscal 28 revenue, and I think it creates a very interesting opportunity in the industry because this actually brings a lot of innovation. So the big picture is, when you think about the mobile business, this has potential to accelerate a lot of feature development of modems, a lot of feature development, and when you think about the role of networks in the age of AI, agentic modems, the network ability to transport tokens, and more important, it sets the foundation to the transition to 6G, which is primarily, you know, an AI kind of, you know, We're excited about that step, and it could have a lot of consequences, I think, beyond the revenue in itself or what we'll do to the mobile industry. It's definitely going to shake up the sector a bit.

Yeah, yeah. Well, that was my next question. What does it mean for incumbent U.S. telecom operators and for traditional telco equipment providers?

Look, I look at this with incredible optimism about the cycle of innovation that it's going to probably unlock in the seller industry. And who knows? As I said to you before, I think we have seen that... Phones change and workloads change, and I think that's probably the beginning of the opportunity. And remember, there are other things that are going to require very high bandwidth and sophisticated modems. It's not only phones, it's going to be robots. It's already cars, and I think that kind of changes the landscape a bit.

So maybe we'll bridge that comment to smartphones, and I think earlier you said it right in terms of, you know, your smartphone business essentially now is de-risked. Apple has moved away, and, you know, from a unit perspective, memory prices, you know, have done what they've done for unit volumes, but we're really at a bottom. So I guess a two-part question. One, what does the recovery look like short-term, medium-term for Qualcomm? And then maybe much more importantly, what does the arrival of an agentic smartphone mean to the future of what will be embedded inside that future smartphone?

Very good. Look, let me break this conversation in two pieces. Let's just talk and unpack the short term. I know because there's a lot of questions, I think, during the earnings call about the short term. So if you look what happened in 26, and I'm talking about the market, the market in general. In 26, and we called this early than I think it was called in the PC just because of the scale of phones and I think the speed of the phone channel. 26 memory prices artificially reduced demand, and we saw the market, the overall market, down in the teams, and it was the overall market down in teams, shipments were down more than that because you also have a reduction of the inventory as now the market's operating at a low level. When we go into 27, We're calling the overall market flat to down because you don't have the inventory dynamics that hit you in 26. So for us, the June quarter was the bottom. And then you have Apple that come out of our model. So when you think about Android, I think Samsung's pretty stable, and we saw the bottom of the rest of the market in June. So now the market is going to remain kind of flat, and as soon as some of the factors change, especially if you have some stability on the memory supply and price, there is demand. There's pent-up demand, I think, for upgrades that is kind of suppressed because of the economics. The interesting question you ask is what's going to happen with the market? I will say that I've started to see, and I alluded to that, the change. I see some AI companies, and let's just use China as an example. I think China, ByteDance was the first one to get approval of the China government to launch an agentic phone. They took over the entire UI of the phone. They started, you know, big OEMs. They don't want to partner with them. They started partnering with ZTE. Now they partner with Lenovo, Motorola. And there's a lot of dynamics in the market about AI companies becoming phone companies. You're going to see some other cloud companies investing in OEMs. And what happens, you change your whole UI. If you look about the harness that you're kind of seeing happening on laptops and enterprises when you just use a harness of AI model, that's happening on phones. The second thing that you see happening is we're getting requirements now from AI companies to think about two operators of the phone. You as an operator of the phone. So you get your smartphone, you have all of your apps, and the apps are designed for you. You go to one app, you go to the other app. Now, another operator of your phone are going to be agents. Think about, like, the Metamuse agent. There's going to be a lot of things going to be done to the cloud, to MCP protocols, but there's going to be some other things. There's going to be some legacy app that you have somewhere, and the agent's going Now go to your phone. The next layer is workloads are shifting. My example of this is... Before smartphones, right, we all are going to use our laptops to do e-commerce. The entire world today will do e-commerce on a phone. Some workloads are going to be more natural for you to talk to your keychain or to your glasses, and as that changes, you start to see a different class of devices that get attached. The phone is not going to go anywhere, but the phone is going to have a role of computing, and the phone is going to have a role of supporting models into the phone working alongside the cloud. Also, it's not a discussion about cloud and edge. Each one, they're both going to go rise. I'll give you a data point that I mentioned, the earnest call. AI companies are asking Qualcomm right now. Can, by 28 timeframe, can I run at least 100 billion parameter model all the time on the phone, inference with an all-day battery life? So that's kind of changing some of the architecture. We, I alluded to this, I think, in our Snapdragon Summit event. We're going to announce it at Mobile World Congress. You're going to see, like, an AI coprocessor for phones using our HBC technology. I think just preparing for the cycle. I think that's the exciting part about the mobile industry. I just cannot predict the time.

Makes sense. Maybe as an aside, a leader in developing and creating a royalty business, could you help me with Meta? I'm a little jealous.

Yes, you should get a royalty rate on me, for sure. But speaking of that, I think our licensing business, we renew, I think, all agreements. I'll say at this point, we never had a situation like this with every single OEM is licensed. I think it's probably one of the most stable times of the license program. I think we're not going to see renewals until post the decade. All of the agreements include 6G, so I think we won't be talking about that business for a long time.

Perfect. Let's switch over to your leadership in automotive and the coming inflection in physical AI. You know, your success in automotive is very impressive. You know, essentially a billion-dollar business five years ago, and it's on track to $10 billion by 27 or 28, making you the largest automotive semiconductor company on earth. So, you know, what are the key drivers of that success, and then perhaps more importantly, taking what you know there to the physical AI world and how you're thinking about that?

Very good. Look, I love this question because I, apples and pineapples, but I will do an analogy about the approach we took to automotive as the approach we're taking to robotics and to the data center. I think we realize that's an industry that is a natural evolution for Qualcomm. We needed to invest with incredible focus. We acquire assets that, you know, were necessary, and we executed on a roadmap. We call it the digital chassis of the car. We didn't invest opportunistic in one product. We invested in the whole platform. And I think that worked very well for us. I think if you look at what's happened on the digital cockpit of the car, I think Qualcomm became a de facto standard. Of course, we're now working with all the Americans, all the Europeans, all the Japanese, all the Koreans, all the Indians, all the Chinese, I think, using some markets. The dynamic became very interesting. We saw an acceleration of silicon content in the car. Our assets of having multimedia, you know, multiple times of compute, CPU, GPU, AI accelerator, and the ability to integrate sensors, low power, have a high-level integration was very successful. So we created a very unique platform to be able to bring assistant driver, assistant driving platforms like level one, level two, to inter-level cars. We have a flex architecture that you can run on the same silicon, the digital cockpit, and the safety of ADAS. And we kind of really went very broad in the channel. So that is driving a lot of scale. I think what you see in the revenue is the pipeline converting to revenues. But it's also accelerating in content. We have now Generation 5 of our Digital Cockpit Silicon. Compared to Generation 3, it's 8x, 8x increase in silicon content, and that has been driving a lot of the ESPs. We're building now full modules that go into the car, which integrates a lot more content in addition of the silicon, and we're gaining share, you know, one socket at a time on processing for autonomy. I think we're running a number of different stacks. We're building our own stack. And I think that market will continue to accelerate for Qualcomm. The next step, we're starting to see in software-defined vehicles, more of the central compute, and a lot of the microcontroller becoming applications on the central compute of the car. And I think we're just at the beginning of Gen AI experiences in the car, the same way that you're gonna see agents. It's natural voice is natural. The cameras on the car to understand context is very natural. So that's going to be another cycle of silicon content in the car. Excellent.

So we've got a few minutes left. So I figured given the audience is largely investors, we can we can move to the investment case for Qualcomm. Your analyst today is targeting $18 plus in earnings by fiscal 29. And for investors who are on the sidelines today and could easily buy an AMD, a Broadcom, a TSMC, what would you want most to emphasize about Qualcomm and why they should buy today?

Well, I think the first comment is, I think there is a case for growth. I think Qualcomm, I will say there is no risk of AI bubble in Qualcomm stock right now. I think if we look at where our multiples have been improving as we diversify the company, but I think Qualcomm. It's still a lot of the valuation is based on some of the core business. I think the investment case for Qualcomm is once we get out of the transition, which we all accept, you know, that we saw this transition that happened in fiscal 26 and 27 of mobile, Apple come off the model, and we've been, we're replacing them once, we're replacing them twice. But when you look, we're very well positioned with markets that are TAMs that are growing. We have built a competitive position in each one of those markets we chose to differentiate. We have, as I said before, the right to play in robotics. If you believe in the Qualcomm automotive story, you believe in the Qualcomm robotics story. And then the last part of it is soon, I think we will be able to demonstrate what we have developed for the data center, and I think that creates an opportunity for Qualcomm to participate in that time as well, and it's all growth.

Excellent. So maybe to close, if we sat down again here in 2030, what would you hope we're talking about?

Hopefully you're talking about how exciting it was that we beat all expectations on our data center projections.

And then what do you want us to focus on into 2035?

It's where we're going to see robotics getting scaled. I think there's no question about how big the market is going to be for robotics, but it's going to take a little bit of time, but not a lot of time. I do the analogy about Robotaxi. We started talking about RoboTaxi a while ago. It took a little bit for the technology to develop. I think it's going to be the same with robotics, and I think that's a very large 10. And as I said in the beginning of this conversation, we're going to be playing robotics a little bit different. We're investing early. So it's going to be, we're not going to be a late entrant like in the data center business.

Excellent.

Well, we've run out of time.

Thank you very much.

Thank you, CJ. I appreciate it. Thank you. Thank you. Very good.

Thank you so much.