QUALCOMM Incorporated (QCOM) Earnings Call Transcript & Summary

Great. Thank you, everyone. We'll get it started here. And we have the pleasure of hosting Qualcomm and Akash Palkhiwala, who's CFO and COO. So Akash thank you for making the conference and sort of holding down the lunch session every year.

Of course.

Thank you for doing that.

I can dive right into the questions, and I know you have to end a bit early. So let's start with auto's strong upside in the quarter. Where is Qualcomm in relation to leveraging the cockpit opportunity relative to initially entering with the telematics opportunity in the market?

Yes. So I think we've made, obviously, a lot of progress. If you think about the 2 changes that are happening within automotive, we think of digitization and electrification. And we're tied to the digitization trend. Right? So that's what kind of pays our bills. And we have 3 -- we think of it as 3 parts to our platform. First is all connectivity technologies. And we've obviously been a leader in that in a long period of time, and we're continuing to hold our position there. The second is digital cockpit. And I would say at this point, any modern digital cockpit, we are by far the leader. And then this is not just in the Chinese ecosystem but Japanese, German, U.S. automakers and Koreans and other parts of the world as well. So it's very, very broad. And I think there are very few sockets we've lost over the last several years. So anything modern cockpit is us. And so this obviously used to be Renesas, NXP market. And as the cockpit changed to something that's required a lot more capable chips, we were the beneficiary of it. The last part of our franchises is obviously ADAS. And I think we've made a tremendous amount of progress there. There's more to come in my mind. The way the cars are changing, there's 2 significant trends that are happening. There's a central compute concept where you have one compute cluster, and it's doing everything that requires compute, and things that used to be microcontrollers are now applications sitting on top of the central compute cluster. And so the Japanese -- the Chinese ecosystem is clearly at the front end of that change. The second thing that's happening is a combining of the cockpit and the ADAS solutions. And the reason why it happens is you want to be able to leverage the input that the ADAS system has from all the sensors, cockpit needs that, too, with a lot of the gen AI use cases that are coming in. The second is Qualcomm can offer 1 piece of silicon that does both at the lower tiers, which is extremely attractive to the OEMs. So I think very good story on auto. I think it's getting stronger, and maybe it's not obvious. I think focus on the short-term revenue forecast, which obviously is doing great. But it's more than that. It's really the trend of change that is happening in cars lends itself to our position getting stronger. So we're very excited about it.

Okay. So in relation to sort of how are customers evaluating Qualcomm and the perspective I ask that from is at the Auto Investor Day that you have posted, the whole vision was that customers start to evaluate Qualcomm as sort of a partner and somebody -- a company that transacts across the board, across digital cockpit, telematics, ADAS. How much of that are you seeing in your pipeline versus still maybe a bit of evaluation of, okay, you're a good telematics partner or you're a good cockpit partner, how is that change reflecting?

Yes, I think the transition has happened and it continues to happen more. I'll give you a great example is gen AI. So if you think about gen AI use cases coming into cockpit, it's a great environment to have a gen AI-based assistant and be able to run models and provide information back to the driver based on things that are happening in the car. Things may be information that's coming in from cameras or other sensors being able to tell driver something. That's a place where if you're an OEM and you want those use cases, you're going to look for someone who has experience in on-device gen AI. And so things that we do in other devices, things that we are doing in PCs, things that we are doing in phones makes us more attractive in automotive. And so I think the concept has changed. It's really not about the OEM going to Tier 1s and 2s, and then Tier 1 and 2 is going through an RFP process. It's really about OEM making a decision of who's going to be my technology partner and then us being picked as that partner.

And then maybe talk about where are you in relation to -- you said ADAS, more to come. But where are you in relation to, you think, in rolling out the entire portfolio that you envision sort of ADAS, you need to have on the ADAS front. Where is Qualcomm's capabilities relative to suppliers like Mobileye or some of the other incumbents in this area?

Yes, I think very strong position. It's just the fact that we are kind of a system provider, not a component provider, puts us in a very good place on ADAS. We obviously came in as the #3 player with Mobileye really being the leader when we entered the business. And now I think at this point, I'd say we're one of the leaders in the market and I think getting stronger as we go forward. Pretty excited about it. We did give a disclosure, I think, at earnings last time where we said $45 billion total design win pipeline. And out of that, approximately 1/3 is ADAS. So that's kind of proof on the numbers behind what I'm saying.

Okay. And so maybe on that front, do you have a revenue target as well for 2026? At least every investor I've talked to after the earnings plan is thinking there's upside to that $4 billion. It would be sort of disappointment if you hit $4 billion and not higher than that. Are there any headwinds that we should consider? I mean, this is in the backdrop of auto landscape that remains pretty challenging. So is -- how much of it should be, okay, there are headwinds that you envision and investors should not be thinking of too much upside to the $4 billion? Or okay, you're on a glide path that clearly is better right now than you thought?

At this point, we're not making any changes to our targets, but we're pretty comfortable with the position we are at. I think we've done better than we had expected in '24. I think we've made a lot of progress towards the target we have set for '26. And so just we'll keep executing. I think there's a lot of good things that's going to come for us in this business going forward.

Okay. So I'll try and ask you that in another way, which is your pace of model launches was a big contributor to the significant revenue uptake. From here on, as you look to the pipeline, what happens to the pace of model launches?

Yes, I think we had -- based on when we launched our different generation of chips. We had a whole set of launches that moved up to our new chips this year in 2024. I think you're going to see kind of a year later, new launches happening as well. And so, again, the market by itself has a cadence of when vehicles are launched. And so we follow that cadence. But I think you should think of us as expanding the number of cars in which our solutions are being adopted every year, and that drives the growth target.

Moving to margins on the autos business itself. You've talked about incremental operating margins being the primary thing to focus on. I mean, is that how we should think about it today as well, that at current revenue as well, it's incremental to your operating margin on the QCT business?

Yes. So the way our entire strategy is based on this premise of we create technology once and then apply it to all these different markets, right? So automotive being one of them, but you think about the PC market, we're, to a very large extent, reusing technology that we create for handsets, right? And then you go to the industrial market, and we reuse the technology that we've created for handsets, PC and automotive, all that goes to industrial. So the whole premise of the company is to reuse. So it's very difficult to talk about any individual business and how the margins play out because it's based on how we allocate the fixed cost. But the way we think about the automotive business is reusing technology from mobile. Of course, it has certain components that are specific to auto that you do have to invest in, which we're already investing at scale. We don't plan to increase the level of investment we have there. And so as we grow from here, that comes down to the bottom line.

Okay. I'll ask you one more on autos before I open it up to the audience. Obviously, autos as a volume -- from a volume perspective, is challenging. But you've seen certain players double down on either being into robotaxi or a fully autonomous solution in the meantime. As much as I've heard you guys talk about ADAS, you haven't talked about where you are in the road map for a fully autonomous solution or a robotaxi solution. How should I think about that?

Yes. Our vision has been that for the next several years, the volume zone is going to be level 2, level 3. And we think the adoption of that is where the highest amount of addressable market is set up for us, and that's where we are focused on. In terms of pure technology capability, there's no reason why we cannot build a chip platform for level 5 as well. But honestly, when we think about how the industry is going to play out, we want to be in the sweet spot of the market and that's where we are focused right now.

Okay. Before I move to the next topic, let me just see if there are any questions in the room here.

So when we think Mobileye, maybe there's kind of an end-to-end approach delivering the full solution versus maybe a more merchant approach maybe just under the chipset. As we think of the 1/3 of the $45 billion pipeline that's on ADAS, how would we think about supporting that for your business, in terms of where that sits on the spectrum?

So the way I would characterize Mobileye is they have an end-to-end approach for a specific vertical of the solution. They're focused on one -- with the vision platform in the car. The way our strategy has been in ADAS is we're -- we have a chip platform that addresses the entire ADAS portfolio. And we also have a stack that is available on its top. We are willing to sell just the chip by itself. We're willing to sell the full stack or components of the stack depending on what the OEM wants. And so our approach has been, we're going to be flexible. The OEMs want to do certain components themselves. That's great. We'll offer our chipset platform for them. If someone is looking for a full solution, we're willing to obviously work with them, BMW being a great example, where we're working with them on a full solution. So we're being flexible about it because I think that's kind of the best way to address the need of the market today. And in the process, we're building the capability to have the chipset portfolio and the full stack.

Let me move to IoT. And Akash, I want to go back to something we have been discussing outside, which is industrial IoT. You've sort of highlighted you'll flesh it out a bit more on the Investor Day. But just maybe give us something in terms of like why is it a big opportunity for Qualcomm?

Yes. So in some ways, industrial IoT today is like auto was 4, 5 years ago. It is a market that's going to go through this transition. A lot of the use cases today are based on microcontrollers wired connectivity. The change to microprocessor wireless connectivity was happening, but when you throw AI in the middle of it, we think there is an acceleration of that change. And so let me give you a few examples of cameras. Doing just vision is not sufficient. If you can combine that with AI, you can make a lot of decisions of -- right at the edge, you don't need to go to the cloud based on what is happening. Second is drones. Drones and robotics really require a lot of the capability that we have in our phone chips. It requires connectivity. It requires great camera. It requires AI. It requires processing, it's a perfect place for us to bring our strength to. It also requires broad temperature ranges. And so everything we're doing in auto can apply to those devices as well. So those are very interesting verticals for us. If you think about AI boxes that can take things that don't have AI, but add AI to them, as an example, a camera network deployed in a store and they're not AI-enabled cameras, but can you add an AI box and do the processing of the feeds at the box and then make decisions on the camera. We think that's a trend that is also happening. So there's a lot of changes that will happen in the market in our minds as a result of the power of AI being used here. And the advantage we have is we have all this AI experience that would come out of phones, that will come out of automotive, that will come out of PC that we can bring to bear. The other kind of industrial PC market is very significant as well. And that market also is going to need AI, and we like our chances of being able to bring that -- bring our PC solution into the industrial market as an alternative to the incumbents.

Okay. Okay. Got it. So let's move to then sort of PC in itself. And you've tried to enter the PC market in the past as well. So talk about why it's different this time around, what's increased the likelihood of success this time?

Yes. So if you think about our product readiness this time around, hopefully, some of you have used our new devices. But if you think about product readiness, we have a really strong chipset. So if you think about the processing performance, I would say, by far, it's the most powerful processor that has gone into a PC. The AI capability is very significant. None of our competitors actually have it. But it's not just about those 2 things. Our CPU is great, our AI capability is great, but the PC is now a communication device. You need to have a great camera. You need to have great audio. You need to have great video. Those are things that we've already done in phones. The other ecosystem does those things on top of a microprocessor. We'll do those things in hardware. So it's very low power. That's why the power performance of the device is incredible. And the quality of the performance, whether it's -- you're doing a video call, you're watching a video or you're doing anything that requires on-device AI, we think we have a very, very strong product. We've been talking about it. Now people have seen it on the device. So I think most people would agree that the best PC chip product is from us. The next thing we need to do, obviously, and which we're in the process of doing is really roll out the devices, roll out the application support, which I think we've just done a tremendous job at in terms of hitting our core market, which is consumers. And then we're working on some of the enterprise things that we need to support over the next several months. So great product. I think there is a willingness and a need from the industry to change. They want to -- the industry needs to change with AI and they need the technologies that we have. So we like our chances. It's a massive market, obviously, in terms of addressable silicon TAM, and we have 0% share today. So as we go in, I think this is a very highly-attractive incremental market for us.

Okay. And going to the application support, like how much more work needs to be done to sort of ensure that compatibility of all the applications with the new chips? I mean, there have been some reports, which I'm sure you've seen as well about sort of older applications not being supported. So when you think about how much more work has to be done and where do you get to a point where enterprises say, okay, this is no longer a decision factor in itself?

Yes. So I think you have to think in 2 parts, consumer and enterprise. The 2 or 3 big things that needed to happen was native support for ARM in Windows, which is true with Windows 11. Windows 10 will get phased out next year. And everything going forward will be Windows 11. So that's a great starting point for us. Second is an emulation layer that would -- is highly performant and competitive to the Mac ecosystem. I think we have delivered that working with Microsoft now. So those 2 were extremely important. And so since then, what we've done is we've picked the top hundreds of applications, and we've tested and ported most of them over from a consumer ecosystem perspective. So today, when you think about consumer ecosystem, outside of very high-end gaming, any use case that you could think of, including casual gaming, I think we're in a very good position. I'd say the job is not 100% done, but we are in a very strong position in terms of having addressed the things we needed to address. From an enterprise perspective, a lot of the core applications, of course, are already addressed because it just comes with the Microsoft ecosystem. There are other applications that we are in the process of porting now. And so as kind of over the next 6 months, we think we'll be in a very good place for a lot of enterprises to be able to launch devices with us.

Okay. Okay. Let me just see if there any questions in the room.

Just a quick question. Given the -- if you can give us some -- kind of twofold. One, if you could give us an update on the NUVIA litigation? And then two, given that litigation, how should I think -- what are you guys doing with respect to RISC-V, particularly -- because if I look at the networking end market, we've had a lot of evolution. We went from PowerPC to MIPS to ARM, it feels like that market could more easily move to another processor architecture. I was curious like how should we think about what Qualcomm is doing? And then how should I think about the mobile market given the dominance of ARM there?

Yes, yes. So on -- on the lawsuit, no update to what we've said before. I think we have an architecture license. We are confident that those rights will be affirmed in the lawsuit. And we feel pretty good about where we've -- the core that we've delivered in PC. Now we're going to bring it to handsets later this year, and we're going to roll it across our entire product line. And so super excited about how that is going to drive a very significant differentiation for us versus our ARM competitors, right? So the lawsuit goes into trial in December this year. But again, we're pretty confident with where we sit, and we'll see what happens. Your second question on RISC-V ecosystem. So we already have several cores in our chip today that is RISC-V. So we have a combination of ARM cores, of course, and the main processors that you -- that we all talk about are ARM cores, but there's a lot of other CPU cores that go in the device and we're transitioning those to RISC-V where applicable. You're right, I think there are certain ecosystems where the application environment is a lot simpler. That's a place where transitioning to RISC-V is much easier to do. I think once you go into an environment where there's a large number of applications working on the architecture, as we just had a PC conversation, that's a harder transition. But Qualcomm 6, we're going to be very supportive of RISC-V. That's something that we're already working on. We're going to keep supporting the expansion of that ecosystem going forward.

What's the right time between to measure success [indiscernible]?

Yes. I think the key question in PCs, and I said this earlier to a group of investors here is really what percent of the market do you think goes to ARM? That's the key question to ask because if it moves to ARM architecture, then you can make your decision on what percent of that you think we get. But our confidence level is, over the next 3 years, a very large portion of the devices will move to the ARM architecture. Just its performance advantages are so significant, I think they're difficult to ignore. And especially in the consumer ecosystem, there's no reason not to transition faster. So we feel pretty good about where we sit. We obviously haven't set like specific targets. You asked the question on 10%. We haven't set those targets clearly, but that's something we'll address at our Investor Day later this year.

This will be my follow-up question, [indiscernible] more on consumer versus enterprise, which takes a little bit longer for the ecosystem to move.

Yes. But if you're asking a 6 to 9-month question, that's true that we start with consumer and then we go into enterprises. And we'll start immediately with enterprises that fit the criteria that we serve the best and then we expand to others. But if this is a 3-year question, then you should think of the addressable market as consumer plus enterprise. It's not just consumer.

Akash, just one question I had on the marketing of the X Elite PCs. I mean, I have one. It's completely changed the way that I travel, great battery life. And yet I haven't touched the Copilot button a single time. I just think it's all-around great PC. But you have a partnership with Microsoft, who is pushing Copilot. How are you thinking about how you're marketing these PCs moving forward? And how much kind of flexibility do you have in terms of how you go to market?

Yes. I think there's 2 parts to this question. One is the Copilot use cases. Second is how do you market the device. So from a marketing perspective, it's been pretty clear the way Microsoft has chosen to market it, right? It's the next-generation PC, it's Copilot+ PC. And every device that has been launched -- I think there's 20-plus devices that have been launched, every single one of them uses our chip. So I think we're in a pretty good position because it aligns us with Copilot+ PCs and even when other players will have those capabilities, we think our performance is so much -- is going to be so much better that we'll have the advantage. The second question is the AI experiences. And so clearly, Microsoft has rolled out a certain number of experiences. There's obviously, a much broader plan of things that they'll roll out going forward, both consumer and enterprise. And so like everyone else, we're eager to get those things out, right? Because those are the things that will further differentiate us beyond just performance and battery life, which is already, I think, we're very good compared to the other PCs, but then you roll -- put on top of that unique AI experiences, and I think that becomes a differentiator for us.

So let me follow up on sort of on the PC market. So Akash, you mentioned you have aspirations not only to sort of be participating in the AI PC market, but the total PC market. Now is that a transition you think driven by AI in that you just sort of wait for the entire PC market to have a high adoption of the AI PCs and that's the transition you're thinking of? Or is there a separate sort of lower-end chipset that then you sort of launch for the non-AI PC market? Like how are you thinking about how do you participate in the entirety of the market?

Yes. I mean, we're obviously starting at the premium part of the market because we want to establish ourselves as a clear performance leader. Then the plan is, and Cristiano mentioned this in his prepared remarks on the earnings call, is we plan to be in the $700-plus PCs by the end of this year. And so very confident about what our chip road map is, how it competes both from a pricing perspective and performance perspective with the incumbents. And that is the heart of the PC market. And so we're very focused on kind of the $700-plus PCs across consumer and enterprise.

Okay. A question that I get often from investors, and I'll still ask it. I don't know if you'll quantify it, but $700 at the low end at the PC market. AI PC is going to go up to $2,000 or more. What is the range of content for Qualcomm?

So you should think of the -- there's an established market and there is the -- if you look at Intel's portfolio, there's i3, i5, i7, i9. We are participating in those tiers at similar price points. So that's how we think about it. We have an established market. We're going in with a better product at similar price ranges.

Any other questions?

I have a question. Sir, the data -- on the data center market. I think your portfolio today if I look at your core foundational technology. I think -- I feel like you guys are leaders when it comes to sort of -- taking that architectural license and really exploiting the performance benefits of ARM. Why not take this portfolio and move it into the data center market? So if you're already getting traction in a PC CPU market, which is a big TAM, the server CPU market is even bigger and in the server market and data center market, there's already penetration of ARM, right? So they sort of laid the groundwork for the team to maybe come in? I'd like to get your thoughts on that.

So I'd agree that our portfolio is extremely relevant to the data center. We have our hands full with the things we're doing, but it's something not yet a part of our plans, but it's something that we'll think about as we go forward. I think the argument that you laid out is consistent with the way we think about it. We just have a couple of other things to do as well.

We have sort of 5 to 10 minutes left. So let me at least change gears here to handsets. How are you thinking about the handset business long term? I mean, from our perspective, it seems like there's always sort of puts and takes with -- in terms of customers that you either win or you lose in a certain year, and that's limited the consistency of that sort of content growth showing up for investors. How are you thinking about sort of it playing out over a longer time frame? Or what can you do to have investors more recognize that as a consistency of underlying content growth even as the numbers overall are a bit more cyclical and volatile?

Yes. So there's obviously a lot in that question. So kind of let me go through like the 5 or 6 things that we think about when we think about the handset market. First is within the handset market, we're at least planning our business for flat units. We have seen, within those flat units, very significant mix shift up. So if you think about devices greater than $400, it's gone from 21% of the market in 2019 to 31% of the market in 2023. So it's a pretty significant ramp up, which is great because we increased -- that increases the content of the market as that shift happens. Then within the premium tier, what has happened is the phones are becoming a lot more capable. And there's always a different driver that is changing the financial opportunity for us. So gen AI obviously, being the latest driver that gets added to the tier of chips and then eventually other tiers and you require a lot more capability that requires memory bandwidth, more memory goes into the device. It makes the device a lot more capable. We benefit from all of that. And so, this is a market where within the same unit profile, you're seeing content increase because of these two reasons, better mix shift and then more capable premium tier chips. The third thing I'll say about the market is there is a pretty good chance that the gen AI drives a change in the size of the market. Today, we don't know it, right? But logic would dictate that, that is a possibility. And so we think of that as an upside opportunity. Again, we are not planning for it, but there's a chance that happens. The other thing that we're keen to see is how the device experience changes, right? Today, we obviously have the app ecosystem experience in the device. We think we're going to go from a smartphone that has AI experiences to an AI phone. And that fundamentally changes how a user interacts with a device and how AI becomes the way you talk to the device. And so everything you want to do, you start there and then that solves whatever app you want to go into or whatever web service you want to go into, it's almost something that's managed by the AI versus managed by the user. And so we think the user experience can change and that obviously means the industry structure would change if that happens, which is another kind of change that's looming in the industry in our minds. From a competitive perspective, obviously, there are 2 large players, us and MediaTek, 2 large merchant players, us and MediaTek. And as we think about our positioning at each of the major OEMs on the Android side, we feel we're in a very good place, right? We have product portfolio lead. We are strong across all tiers. We have new technologies, gen AI, our custom CPU coming and waterfalling across all tiers. So we like the hand we have in front of us. And then maybe the final thing I'll say is we did this RF front-end acquisition. Now we have modem plus RF front-end. China is deploying 5G advanced right now, this year. So next year, all devices will have to now support 5G advances, and 5G advance and new features that come in with it. And then at the end of the decade, we are looking at a transition in that interface that is primarily driven by AI. So over the last several years, 6G is being designed with AI at the center of it, and we're seeing that play out as well. So there's a lot going on in the phone market, in my mind, all positive things. When you look at 6-month windows, you might come up with certain positive things, certain negative things. But when you kind of step back and look at the trajectory of the business for a longer period of time, we think we're in a great position. And again, remember that this is the business where we create the technology, we apply it to all these new areas I just talked about. And so it's almost you're expanding the SAM of the R&D that goes into the handset business by leveraging it at all these other places.

One thing I've seen you hesitate around is sort of predict the AI smartphone adoption related to AI PCs where you've been a bit more open about sort of predicting it. Underlying driver of why more hesitation on predicting the AI smartphone adoption cycle?

Yes. So I'd say AI PCs with Microsoft and their clear vision and their plans is really kind of what gives us the confidence there. And in a lot of ways, we are projecting what their belief is and the industry's belief is. And of course, we benefit from it. And then again, the way our PC strategy works is we're going to make the best chip that the Windows ecosystem has if you don't do AI. And if you do AI, it's even better, right? Like so that's kind of our strategy going into it. On the phone side, I'd say, we've gone from like 5 experiences to 15, 20 experiences. First half of '24, there were maybe 5 or 6 experiences on devices. We've gone to 15-20. We're going to go to 50 next year. We feel pretty confident that over time, that is going to happen, right? This is going to transition over to an AI smartphone. The question is, what does that do to the scale of the business which I don't know at this point. And while I have a personal belief that it should drive better replacement rate, more expensive devices, we just want that to play out. I don't want to plan my business based on it. We think of it as upside.

Any questions. Tony?

It's not necessarily smartphone, but somewhat related. So you obviously had lots of initial success until the market basically cratered on VR, but Meta clearly is all-in on AR, and they've had some initials success, AR glasses and AR plus AI and this could be, obviously, a very large market conceptually, and you have your compute for that in the wireless. But there's still lots of technical challenges to pull off a full AR glasses. And so to enable that further, could you -- do you envision going deeper in the stack like the optics and everything else to accelerate pull forward of AR glasses? Because then you're not sitting here talking about flat units, you can have a market that could drive both sides, but like you're just sitting there with your RF and compute. But the problem is not just RF and compute, it's everything else around the AR. So do you want to get deeper into that stack and help Meta get out there?

Yes, great question. So we are absolute believers in the AR/VR market, and we absolutely are aligned with the view that AI changes the opportunity for AR. And if you think about the Meta glasses uses us, we have basically every major OEM, Google, Samsung, Chinese OEMs, every single one uses our solution. So we're in a very good place. It's not just about connectivity and processing, right? Obviously, camera is super important. All the on-device gen AI capabilities that we just discussed are very important. You think of these as companion to your watch or companion to your phone, where you can get a query to your glasses, send it to your phone and run the query on the phone, I think, is a very reasonable use case that could happen because the context about the user that the device has and the glasses has is tremendous. Always-on camera with computer vision, very, very important because you need to make sense of everything you're seeing. And so being able to just do simple use cases like, who am I looking at? And what am I looking at. How many calories in this set of vegetables that or this drink, Starbucks drink that I'm looking at. There is just so much that can be done with AI use cases. Translation is a great one. I was just in China, and I tried these glasses that was giving me live translation to someone talking in Chinese to me. And I could see live translation happening in the glasses. So very much in the middle of that use case, that device. And we're doing a lot more than process, still within the chip framework and the software framework, but we're doing a lot more than just kind of processing and connectivity there. VR has a set of problems that need to be resolved on the optic side and other areas. So while we're also the chipset of choice there, our view, our optimism on the AR market is much higher at this point.

So I'll end it with one question for you, Akash, on capital allocation, which is you -- at the end of the day, you have a significant exposure to a cyclical end market. And when you think about your investor base, that probably looks for more reassurance of more of a consistency of earnings growth, why not double down on the capital allocation front to provide that? How are you thinking about capital allocation between buybacks and dividends, but why not essentially sort of ensure a lot of the earnings growth through the buyback using your balance sheet?

Well, I think we are doing our part in that. We've obviously laid out a strategy and we're executing to it, and we've been consistent and transparent about it. And that's, I think, as much as we can do. In terms of being a lot more aggressive with buybacks, our general view is that this is an industry where a strong balance sheet is important. We have all these growth opportunities that we discussed. We talked about auto, PC, AR/VR, industrial. There was a question on data center. So there's a lot of opportunities in front of us. And as you think about the relevance of our technology to those areas, I think going into that conversation with a strong balance sheet is important for us. So that's the balance we're trying to strike.

Okay. Great. I'll wrap it up there. I know you have to head over to the airport. So thank you, everyone. Akash, thanks for coming to our conference. Thank you.

Thank you.