Jim McGregor here from Curious Research at CES 2026. I am joined by Nakul Duggal who is the Senior Vice President and General Manager of the Automotive and Cloud Computing group at Qualcomm.
And automotive. How about we just call it all the cool things? Sure, I'll take that. Good to see you.
It's great to see you. And we're joined today by the newest Qualcomm employee, Paul.
Hello, Paul. So, this is great. You guys made some great announcements here at CES, especially around IoT. You got new processors for IoT, the Q processors, and you got the IQ10 specifically for robotics, not to mention a lot of other announcements. Qualcomm has done a lot over the past just 2 years in really expanding its IoT presence with a lot of acquisitions. I'm trying to remember them all now.
Arduino. Yes. So I mean what's all this mean? How does all this come together?
You know, so we've been on this trajectory to transform the company. We've been known as a smartphone company for a long time. That's been our whole business. But I think we started to focus on automotive about a decade ago or so. And the approach that we took was to really think about ourselves as a technology and a systems platform company and then really envision what it means for a platform like a car to be a car of the future. Now as we took on that journey, a lot of technology trends came into the picture from the point of view of the car, software defined vehicles, central compute, but then AI became a part of everything and that allowed us to really be able to think about what does it mean for us to take our technologies and build an advanced compute platform architecture, almost a framework in terms of how you think about all these things. Now, we were very fortunate because the time that we picked when we were starting to focus on automotive gave us a pretty blank sheet of paper to think about what these platforms will look like. And it could not have been better because we've been able to really change not just the portfolio, but really the thinking and the culture in the company in terms of what does it mean to solve problems in the industrial space, in the robotic space, in the automotive space. And it is very different than building a hardware product that is running some basic software on it. It is about continuous software evolution. It's about training. It's about taking what you train and deploying it. It's about being conversant with models, all kinds of models, large language models, vision language models, computer vision, audio. And it all kind of hangs together because you're talking about transformation in the physical space which in my mind we are still in the very early days of realizing but it's been a fantastic journey because we get to touch everything from technology to the instantiation of technology into a full product platform, the software, the AI and then work with customers of all shapes and sizes.
Well, and that's a great point. I mean, we talk about the processors, we talk about the IQ10, which is powering Paul here. However, there's a lot that goes into that with the tools, the software, the ecosystem development, all the things you need around that environment. 2025 really saw an interest perk up in robotics and now 2026 things are taking off very rapidly. So, very excited to see more of these and what's coming. But I know that those aren't the only announcements you have. So we want to talk about, you mentioned automotive. You guys have had a lot of automotive announcements and you're a powerhouse now in automotive.
Yeah. You know, I think automotive has really been a fantastic story for us. And it really comes down to putting yourself in the shoes of the automaker and we work with every automaker in every corner of the planet and especially when you work with automakers that start from a blank sheet of paper that don't have a lot of legacy that have the ability to really rethink what nowadays what has been called the AI defined vehicle because it's no longer software defined it's AI defined, what does that actually mean and we are learning very quickly what it means for a lot of these trends to get deployed at scale at the edge. Right? So one thing that I think is very important for folks to understand is edge AI is being deployed at scale in automotive both agentic AI inside the cockpit and physical AI as ADAS. Now we have built a compute road map that is super scalable. We go all the way down to the very entry tier of vehicles which are all and will always continue to be very cost sensitive all the way to the luxury tiers where we announced our cockpit elite and our ride elite platforms a year and a half ago. We have commercialized already in 11 months from silicon on dock to the car being commercialized with 8797. That's what we announced with Leap Motors here at CES. In 26 we will see a dozen additional customers commercialized on the 8797. So the pace at which new technology is being embraced by the automotive industry. But not only that, the rate at which new platforms are being deployed is a sign of how rapidly people are starting to think about what incrementally new features do I want to deploy, what software do I want to reuse, where do I bring AI in, where am I differentiating on new features that the hardware is introducing. And because the automotive ecosystem is so global, whatever is happening in China is being seen by every global automaker. They are learning there, they're deploying there, they're taking those learnings and then they bring them back to their home countries. So, we are super excited to be part of this flywheel because it allows us to just run very fast.
So, I mean, you announced at UI Mobility the software partnership actually come to fruition with BMW, new software platform. I know that you've got announcements here at CES including a new partnership with Volkswagen.
So I mean you guys just seem to be marching forward and the fact that you're making this kind of an open ecosystem for everyone to join. Should we expect the same thing out of everything you're doing with robotics and industrial automation and all these other segments?
Yeah, let me maybe connect the dots a little bit for you. So you know what was very successful in automotive was us basically realizing number one that we want to build platforms that will run applications that we don't quite know of or cannot envision. Therefore we have to have the building blocks in place. We might pick a certain area that we want to do something more in but that doesn't mean that we don't want others to also go do that. That created this opportunity for us to think about really an endless set of applications, models, use cases that should run on our platforms in a seamless way. Now when we pivoted towards industrial, I took over the industrial business about 2 years ago. What became pretty clear was that we have to have the same recipe book, but do we have all of the tools in place? Do we have the culture in place? Do we have the ecosystem access in place? So we acquired Arduino and Edge Impulse because we needed to be developer friendly. And the part that is developer friendly there is we have all the chips in the world that you need. We have the software. But how do you reduce the distance between what a developer wants to do on our platforms? So these ecosystems and these tool chains have allowed us to be able to go bring that same capability. Now if I think about robotics, what I really like about robotics is that this is physical AI and the use cases that you will see in robotics haven't been imagined yet because physical AI is a new thing. There is nothing like physical AI like we did not call ADAS physical AI but in this case it is about teaching a robot how to perform a certain task and then that becomes a skill. Once it becomes a skill the important part is how portable is that skill. So this is a simple pick and place application. If I wanted to make this work in a retail environment where you're picking and placing boxes on a shelf or you were in an assembly line and you were picking and placing packages of various shapes and sizes, it's about how quickly can you train it. We look at it in two ways. One, what is the infrastructure that you have to have in place to be able to go scale this up, right? So we have the chips, we have the training recipes, we have the development platforms, we have the ability to be able to run software models really on any kind of silicon and then the training of the silicon. Now if you think about the verticals that we are going after it's very interesting because we are talking to a large number of industries who have a net new need to be able to replace labor with automation because they're not able to hire people to go to those jobs. Security and surveillance. Agriculture. Agriculture, hospitality, and healthcare. If you can build a platform that is a generic platform that can add more skills over time, now what you've done is to create something that you can keep investing in and it will keep paying dividends because you're basically increasing the skill level of the platform. So what is very exciting about this is that there is really no end to the number of ecosystems that you can enter because many ecosystems have never really thought about automation with a robot because there was no solution available. On the other hand, there are several ecosystems, warehouses that have had some level of automation or factories that have had rules based automation that have the ability to bring more automation in because it is safer because you actually have a use case for it. What I really like about it is that the segmentation of the markets is very early.
