


How AI demand is reshaping mobile networks and data centers, Nokia’s “picks and shovels” role (IP networking + optical connectivity), and what AI RAN and 6G could enable (sensing, physical AI, edge intelligence).
Justin Hotard, Nokia CEO (appointed last year); previously led Intel’s data center and AI group. Background includes chips startups, supercomputers, cloud, and AI infrastructure.
AI is “still very early” yet already about a third of internet traffic; persistent constraints mean “we’re not overbuilding” (if customers could build 2x faster, they would). Networks must become “sense, plan, act” and more secure for trusted, time-sensitive decisions. 6G is less about consumer speed and more about software-driven, differentiated services (AI RAN) and avoiding “forklift” upgrades.
AI-era data center interconnect (IP + optical) enabling multi-location agentic workloads; physical AI use cases like autonomous cars sharing brake/incident actions by location and time; optical’s power efficiency vs copper (longer reach without re-amplification); Nokia scaling optical capacity via laser/indium phosphide tech and an NXP fab partnership.
Written by AI. May contain mistakes.
Discussion about Nokia's transformation from mobile phones to digital infrastructure.
“Its products had been core to the mobile infrastructure that we all rely on for our data and calls.”
Justin outlines Nokia's current focus on AI-driven digital infrastructure.
“So this is going to be an exciting conversation because a lot of people remember Nokia and the brand and the heyday of Nokia with the mobile phones.”
Exploration of Nokia's contributions to networking and optical technology in data centers.
“So if you hear about AI factories, that's networking that's providing that.”
Discussion on the current and future demand for AI infrastructure in the market.
“Is demand or is building happening too far ahead of demand?”
Analysis of the value AI brings and the constraints affecting its development.
“And then the other thing I would say is when you look at what's happened in the last couple of years, we've been in a situation of persistent constraint.”
Examining the potential for Nokia in physical AI applications like autonomous vehicles.
“So that's almost where your more traditional telecommunications business now interacts with kind of the newer data center business.”
Discussion on how AI will transform mobile networks and concepts like AI RAN.
“And so what kind of opportunity for Nokia is there in physical AI, this idea of robotics or autonomous cars or any kind of physical thing that is using AI?”
Explore how AI is transforming mobile networks and the concept of AI RAN.
“When we think about AI native networks for mobile, that's what we see.”
Discussion on the evolution of mobile networks from 5G to 6G and its implications.
“But what are the current discussions around 6G?”
Insight into how software will drive the next generation of mobile networks.
“We're partnering with other players in the space.”
Examine the trends in data center buildouts and the role of AI in connectivity.
“There have been these big announcements regarding huge data center buildouts, a certain amount of gigawatts here and a certain amount of gigawatts here as well.”
Explore how optical technology enhances power efficiency in data centers.
“Everyone talks about the fact that these data centers can consume a lot of electricity and resource, etc.”
Learn about the significance of lasers and semiconductors in optical communications.
“Can you just lay out kind of the difference between optical and copper?”
Discuss Europe's technological landscape and the importance of connectivity for AI.
“And you're investing in capacity around lasers, around your own semiconductors for this solution, right?”
Understand the existential need for European industries to modernize with AI.
“And I think it's actually a pretty existential issue for Europe because the reality is the internet was largely about media.”
Analyze the growing concerns surrounding AI safety and industry responses.
“The AI safety debate has burst back into the public view.”
Explore the safety concerns regarding AI and its implications for mobile networks.
“which is a very key point, will keep pushing them faster.”
Discuss the current demand for infrastructure and the potential impacts of slowing AI advancements.
“I mean, you're basically going to damage the company.”
Examine the relationship between data centers and local communities, addressing concerns and benefits.
“Well, I think this gets, unfortunately, there's probably some bad examples that create a lot of news.”
Learn about Nokia's initiatives in space communication and potential future applications.
“I mean, when we spoke, I think, in February this year at Mobile World Congress, which is usually where we meet, I asked you about space.”
Investigate the feasibility and potential of data centers in space for improved connectivity.
“And so I'm really excited about where we're headed there.”
Discuss Nokia's focus areas for growth and the possibility of returning to the mobile phone market.
“From our part, we want to be ready to make sure that where we can be relevant in that solution, we're delivering the innovation and the technology needed to make sure it can happen.”

Get the gist, explore an idea or ask about something said.
On The Tech Download, what are the key points of this episode?
Automatic transcript. May contain errors.
0:01Hello, I'm Arjun Kharpal and welcome to The Tech Download. And on today's episode, my guest talks about the current state of AI demand. If we could build 2x faster, our customers could build 2x faster, they probably would. The future of mobile networks with 6G. I think there's a view of end of the decade. I think we're going to see 6G services faster than that. And his view on the potential of Nokia's business in space. We think it could be massive.
0:29So who had a Nokia phone? Who remembers the game Snake? I think a lot of people listening will remember this because for a generation, Nokia was the biggest mobile phone player in the world. But it lost its way, of course, as Apple's iPhone and Google's Android reshaped the smartphone market, and Nokia struggled to adapt. Its products had been core to the mobile infrastructure that we all rely on for our data and calls. But now Nokia is using that expertise to write a new chapter. as an infrastructure company sitting behind the AI boom. Its technology helps connect the fast data centers powering artificial intelligence from the networking equipment inside of them to the optical systems that link them all together.
1:07And it's pushing to bring AI deeper into mobile networks too. Leading that pivot is Justin Hotard who became Nokia's CEO last year after running Intel's data center and AI group. So he's got this interesting history in the data center business and he joins Nokia with a background at the intersection of chips startups, supercomputers, cloud, and AI infrastructures. Investors have been cheering the shift. Nokia shares are on a tear as demand from AI and cloud customers accelerates, and the company positions itself as a picks and shovels player in the AI build-out.
1:43Justin, it's great to have you here on The Tech Download. Great to be here, Arjun. Thanks for having me. So this is going to be an exciting conversation because a lot of people remember Nokia and the brand and the heyday of Nokia with the mobile phones. And everyone remembers having a Nokia mobile phone. But the business these days is something so different and something so interesting because Nokia has found itself at the center of this kind of AI boom that's happening right now. What is Nokia's business today? How would you describe it? Look, today we're a digital infrastructure provider. We provide technology across all of the core elements that are enabled by the AI super cycle.
2:18data center, you know, to the edge, ultimately with what we're building in AI RAN, which we've made great progress on over the last few months in our partnership with NVIDIA. You know, that's our journey, enabling everything from, you know, from agentic AI to physical AI to powering the service providers, the hyperscalers, and even delivering some new innovation around new applications and services that weren't previously possible. The data center story is a fascinating one because that's been a big pivot for yourself under your leadership at Nokia. Typically, Nokia has been this company that has been selling very key equipment for the mobile networks you and I use and everyone uses to power their 5G and all of their services and apps and connectivity as well.
3:06The data center story is interesting. Can you just lay out a little bit about where you play in the data center as we talk about these massive AI gigafactories that are being built? Yeah, absolutely. So there's two things, Arjun, that we do to keep it simple. One is all of these data centers, these AI factories, they have a ton of networking equipment, IP networking equipment. And that's something that Nokia has been providing for years. It largely came through the acquisition of Alcatel, but Nokia is built on that, continued to invest. And that layer is often talked about within the compute infrastructure.
3:42But as importantly, it's what provides all the connectivity for scaling. So if you hear about AI factories, that's networking that's providing that. Or if you hear about data center connectivity, we call data center interconnect, that's also enabled by networking. We call it wide area network routing. The other part of that that we provide is the optical network. And that's actually the layer that connects these big networks together, these big data centers together. And from that perspective, we're one of the few players in the world that has advanced optical technology connecting these data centers.
4:15And what's changing so rapidly is the importance of speed and bandwidth and latency and, of course, security of these networks. So connecting these data centers in real time because more and more as AI becomes prevalent in software development, in agentic bots that are running parts of networks works for our traditional telco customers to financial services, other applications, other vertical applications, to what's coming with physical AI, autonomous vehicles. It's not enough to just have the AI model give you an answer. You need it to give you the right answer at the right time so that you can make the right decision or take the right action.
4:57And that means more bandwidth, more connectivity. And all of that is powered by Nokia optical and IP networks. So just to paint a bit of a picture as well, Justin, when these companies are building these data centers, they're installing these huge racks of servers containing all the chips that people have heard of from companies like NVIDIA to AMD and others, et cetera. Obviously, these racks need to be connected in some way, but the data centers themselves need to also be connected to other data centers elsewhere. This is all part of what you do, right? That's right. And that's that connection of IP networks and optical.
5:28And to your point, Arjun, if you think about in the internet era, like I'd call a webpage, or even if I'm streaming a video, right? That's largely going to one data center. There's maybe something at the edge that's helping buffer or serve that internet access. But it's usually pretty monolithic. I go up and hit one site. Maybe I go hit a couple of places because I'm accessing some data. In the AI era, I've got agents that need to take actions on processing and inferencing data. They need to then make a decision, take an action. They may kick off another query. All of that requires multiple access points.
6:04And so that means these data centers need to be very connected to each other. And that's something that we hadn't really seen in the internet era. That's a big part of what we're seeing today in data center demand. Justin, you're in a perfect position to talk to us about demand. And it's something, I guess, the investment community is looking at and saying, well, we're seeing billions and billions, hundreds of billions of dollars in spending by companies going towards building out AI infrastructure. Is the demand there over the long term? We know this build-out is happening. Is demand or is building happening too far ahead of demand?
6:37These are all sort of debates people are having right now. You've seen, I mean, phenomenal growth, I think, of that AI and cloud business of yours just in the second quarter of this year. Just give us a picture right now of what demand is like. Is the building happening at the right speed, too slow, too fast? Yeah, I think first of all, if you look at where we are, the demand we have is still very early. I mean, you know, again, AI is quite early. So, you know, yet it's quite early, but it's a third of the internet traffic we see already. Wow. So it's already massive in terms of the amount of data it's consuming.
7:10It's much richer than what we've seen in prior. And it reminds me, we talked about Nokia on the phone days. It's a little bit of the same experience, right? Looking back at history of, you know, the internet was so much more data than what we had seen with SMS or, you know, mobile phones. So we're seeing that step function of data. I think we'll see the same thing with AI. And that's why we feel pretty confident we're at the beginning of this cycle. Having said that, if you look at the value that AI is delivering, I mean, the value for companies like us in terms of accelerating software development, providing greater intelligence and insight on decisions we're making, the value is also there as well.
7:48And I think what's different about the AI super cycle is this is much more of a business or enterprise driven cycle at the beginning versus a consumer driven. We all remember the dot-com era being very much about pets.com or consumer information. But I wouldn't discount that. So the business value is there. And then the other thing I would say is when you look at what's happened in the last couple of years, we've been in a situation of persistent constraint. So we've had energy constraint. We've had silicon constraint. We've got supply constraint across other parts of the curve and the ecosystem.
8:27So from that perspective, I don't think you can say in any manner we're overbuilding today because reality is that if we could build 2x faster, our customers could build 2x faster, they probably would. And so that gives me confidence that we're still in the early days. Now, we all know that we're not going to grow it at this kind of rate forever, and we will see digestion periods. But I think when you look, the analogy I tend to look back to is not the internet boom, but more the beginning of the cloud build, where you would see these periods of growth and digestion and then growth and digestion.
9:01I think we're still in that first phase of growth, and at some point we'll see digestion. But right now, we've got a lot of constraints preventing that digestion period. I think that's what I wanted to sort of understand as well. how do you continue to benefit from this development of technology? Because if I look at, you know, the build out of mobile networks, for example, and, you know, the value began to accrue to other companies, smartphone makers, the app makers, those services that were being built on top of the network infrastructure, Nokia and others effectively helped build out. So are you worried something like that might happen in this kind of AI pathway?
9:39Or do you feel that you're going to be able to take a lot more advantage of the value that's being accrued in the ecosystem this time around? Yeah, we did 4G, we did 5G, but we didn't innovate in anything that was differentiated in new technology. If I look at what we're just starting to do in AI-Ran, Arjun, what I see there is very different innovation. We're talking about sensing, hyper-accurate location positioning, providing that data as a service. So we're providing new tools. Now, Ultimately, our customers who buy the infrastructure from us need to be able to monetize that and deploy those services.
10:13But there's no question in my mind that the data is going to continue to grow at the edge. The need for processing at the edge is going to continue to grow. And that means there's an opportunity for us to continue to innovate. So that's almost where your more traditional telecommunications business now interacts with kind of the newer data center business. because, I mean, we're using various apps, AI apps on our phone, whether it's Claude or ChatGPT or Gemini, whatever it might be, that right now requires mobile connectivity, you know, in the form of broadband or 5G to some extent. That's right.
10:45And today you can think of that connectivity, Arjun, as largely running, it's running like a data query, but we're already seeing elements of that connectivity being very different because in the internet era, I would, you know, largely what I do is I send a request and I get data down, but it's very download heavy. It's almost 12 to 1 ratio in mobile networks. If you think about what we're seeing today with AI workloads, it's 4 to 1. So it's still download dependent, but at a much lower rate. Why? Because some of those queries, I do this quite a bit in Finland. I take a picture of something, I send it up and I ask, you know, tell me what that is.
11:17Or, you know, I ask a question about where I can find something, right? And that's images. And sometimes it's audio for processing. And then you think about what's coming, what we're seeing with physical AI. The early autonomous vehicles, they're going to process things locally. They're going to send images up. Robots, same thing. You can't have all of that processing be local. There'd just be too much weight on the device, too much battery consumption. So it has to happen at the near edge. And that's the opportunity we see for networks and certainly for telcos, should they seize it, is to deliver these differentiated services.
11:53And this is why we talk about the network being a part of the AI fabric in the future. Yeah, that's so interesting. I want to deal with more of the data center conversation. But since you mentioned the edge, I think it's good we talk about it now because this is the idea. And just to, again, spell it out for our listeners and viewers, the idea of processing any kind of AI on another device. That could be your phone. That could be your tablet. That could be your laptop. It could also be a driverless car or a robot or some other kind of device as well. And so what kind of opportunity for Nokia is there in physical AI, this idea of robotics or autonomous cars or any kind of physical thing that is using AI?
12:31Yeah. So first of all, if you think about, let's take one of those examples. Let's start with your example, Arjun, of cars. If you think about those cars, those cars today will have arguably a lot of compute power. They'll process some things locally. But today they operate as independent vehicles. And what happens in the future is you want those cars to start to talk to each other. There's a traffic incident here. There's an accident here. Oh, I'm slamming on the brakes because someone's crossing the street that I didn't expect. You want the autonomous vehicle, three cars back to know that because then it can start to slow immediately.
13:08Now, think about that action. That action is the car that's slamming on its brakes needs to make sure it's slamming on its brakes, needs to communicate that that's happening. Now, when it communicates that's happening back up to the network, you don't want to send just an image or, you know, here's all the data that this car is doing. You want to transmit to that car, this is happening. This vehicle is taking this action in this location. Now, that little activity requires the action that's being taken by the vehicle, knowledge of the locations. You don't want to send that to a car that's five miles down the road because it'll slow down in the wrong place.
13:44So you want to make sure it's the relevant location, and then it's very time sensitive. So you think about just that example. It's one minor use case where the network needs to be aware. The network actually has to have the intelligence to say, I'm going to share this out to these vehicles, which means the network needs to see location, needs to see time, and it needs to have some awareness of the application and the service it's providing. When we think about AI native networks for mobile, that's what we see. We see these radio towers becoming much more intelligent. And that's just the beginning.
14:15That applies to every element of physical AI. So what does this mean then for future mobile networks? Because there are a couple of terms, maybe our audience hadn't heard of before, AI RAN or radio access network. What is that? And why would that be important in the scenarios you're describing? Yeah, what we see with AI RAN and why we need a computer at the, there's always been a computer at the base of the radio tower. It's been that way since, really, since 2G. But that computer has been very focused on one thing. That computer has been very focused on delivering our connection, voice calls, data calls.
14:53Now, when we think about this world where the computer needs to deliver tokens, intelligence, the computer, first of all, needs to have the flexibility to make sure it can deliver the intelligence it needs. That might mean actually doing some processing. It might mean sharing additional information back up to an edge inferencing environment where that token can get processed so the right action can be taken. So it means these networks need to become networks that sense, plan, and act. It's very different than I'm just transmitting a packet. I'm just transmitting a data packet, right? And the other element of this that becomes important is these networks also have to become inherently much more secure.
15:32Today, the radio network is quite encrypted, but we need to make sure that when that computer starts to become a part of the intelligence network, we need to make sure that it's seeing information that's trusted, it can verify. You know, that packet that came over from that vehicle is a trusted vehicle. I know it's in this place. I can validate it. So we need to add a layer of capability so we can have that kind of intelligence. And that's what's different about AI RAN. And that's why we see the radio network being so different in the future, even if the fundamental thing, the radio, isn't going to change that much.
16:05Right. And where does, you know, we talk about 2G, 3G, 4G, 5G, the industry, your industry is now talking about 6G. I guess from a consumer perspective we saw a huge jump when it came to mobile from 3g to 4g in terms of speed and underpinning all of the the apps we use streaming etc 5g for consumers anyway bought some faster speeds um but the promise was that this would be a really b2b kind of thing for enterprise and it sort of hasn't quite panned out in that way but the industry is talking about 6g and i know the telecom industry works many many years in advance before a technology actually comes to pass but But what are the current discussions around 6G?
16:45You know, what is the purpose of it versus, say, 5G? Yeah, there's two things that I think traditionally push together that I would separate in these Gs. So one is the Gs become a discussion around new spectrum, new frequency. Right, yeah. Which is ultimately for the industry a way of getting more capacity. Executive Decisions is the new podcast from CNBC, where I ask powerful leaders about their decisions that changed everything. I'm Steve Sedgwick. Here's the CEO of the London Stock Exchange Group, Julia Hoggett. There is no justification whatsoever for not doing the right thing, even if the right thing is hard to do.
17:22Which also means that if the context changes or the information changes, it is absolutely fine to also change your mind. That's Executive Decisions with me, Steve Sedgwick. Get it wherever you're listening to this.
17:38One of Nokia's core businesses is the equipment that helps mobile operators build their networks, the systems that let us make calls, send messages, and use the internet on our phones. Now, we know those networks by their generation names, 2G, 3G, 4G, and now 5G. Now, traditionally, each new generation has meant operators buying significant new hardware from companies such as Nokia, Ericsson and Huawei. But the next shift is not only about a new G or faster speeds, it's actually about making networks far more software driven and intelligent. Now, that doesn't mean the hardware disappears. Operators will still need more capacity, new spectrum and in some cases new equipment.
18:20But more of the network's capability can be upgraded through software. That matters in the AI era. Now, AI applications create more complex, time-sensitive traffic. And networks will need to decide dynamically, automatically, what data matters, where it should go, how quickly it needs to get there. That could become crucial for things like driverless cars, robotics, and emergency services. That is where AI RAN, which stands for AI Radio Access Networks, come in, as Justin Hotard explains.
18:53So sometimes when you hear people talking about 6G, they're talking about new spectrum. And with all these things we just talked about, there's going to be more data, which means more bandwidth is needed and more spectrum. We see that coming. In 6G, in terms of technology, we see this being much more around machine-to-machine communications, these sensing applications and services. That's why we partnered early on with NVIDIA, not just because of their incredible computing capabilities, but their understanding of full-stack AI and the services they can build. but it's not just NVIDIA. We're partnering with other players in the space.
19:28But the key thing for me in this is this is really about software. And when I talk to a lot of our customers, that's the shift they would like to see as well. The past in the industry has been every one of these GEs requires a huge forklift hardware upgrade. And the reality is most of the cost of that is changing radios, even if the radio itself doesn't really need to change. And so the shift that we're making with AI RAN is also to say, look, one, we don't want you to switch radios unless there's some reason to do it. You know, the radio's 20 years old. It needs to be modernized. You're adding this new frequency I talked about before, a new spectrum.
20:06Okay, you need a new radio. That makes sense. But we want to have interoperability at the radio layer because that allows customers to make software upgrades. And there's some customers that we see that really understand this, that are visionary. Some of them are our partners in early AI RAND trials, and they understand this subtle difference. And what that allows the industry to do is not to worry so much about doing a 2G, 3G, 4G, 5G jump, but rather let's just build the services. And I think what we'll see differently with 6G is we'll probably see the little logo on our phones that goes from 5G to 6G, but more importantly, I think what we'll see is different services.
20:43And I think if you look at 5G, the reality is for most customers, maybe the bandwidth got a little bit better, but there really wasn't anything different. And I think in 6G, what you'll see is, maybe we won't see it as consumers, but what you'll see is, wow, this network can now handle sensing. It can now handle physical AI. And that's where we see the opportunity in 6G. How that plays out, I think is probably going to be a little bit different than what we've seen in the past. And I think it needs to be, because the reality, when you look at the industry is the industry hasn't been very healthy when it's gone through these big forklift upgrades.
21:15And we see that as a partner to our customers. We also see it from our customers' perspective, because when you look at their return on capital, it hasn't been there. Yeah. Do you think that could change with 6G? What are potential ways that could change? Is it just underpinning new applications? Yeah, I think it's the opportunity for the industry to charge for these applications differently, charge for differentiated levels of service. It's also that the business model evolves because a lot of what we're talking about is not a consumer device in a subscription. It's really an enterprise device.
21:47And we have some early indications of that with things like smart glasses and even autonomous vehicles where these connections are mission critical for the devices. But it's also the opportunity for the operators to charge differently. But the other key thing is don't force this big capital bubble because that means that they don't have to make these huge investments. Right. So that could be a big differentiator versus previous generations of connectivity is i mean one of the other big discussions with over over the last few years has been the fact that you know telcos have built out this infrastructure um of fast connectivity that then internet companies have effectively built on top of right and there was this whole discussion oh should companies like netflix or google or whoever pay for this infrastructure given that all their services are being delivered to consumers over that infrastructure as well.
22:37So that became a discussion around how you should charge for it, who should pay for it, et cetera. Do you expect something similar to happen with kind of 6G, given again, it's those networks, as you've described, that we're going to be relied on by companies who are developing AI models or AI services, physical AI as well? Look, I think the opportunity for there to be a different story here is there. How it will play out is probably a little bit harder for me to predict. My focus is on making sure that we are delivering the cutting edge technologies and capabilities that allow this innovation to happen at the edge and then to let the market take shape.
23:12But I think the opportunity is there. The key thing for what I tell my customers and my partners all the time is, but you have to be a leader in innovation. And if you're not differentiating on the performance of your network and delivering these services, you're probably not going to be successful. And I think, you know, when you actually go back and look at what's going on in the industry, even with the challenges that you described in the past, the companies that have built great networks that have differentiated on performance have delivered outsized returns. So the opportunity even to monetize the services today are there, but you have to invest in the network.
23:46And I think that's going to be the biggest. When we look ahead, that's the biggest, you know, differentiator that we see in the short term is invest in the network. Because it's not everything we just talked about is about new features and new services. It's not just about that. It's also about trust and security, right? You know, there's no way I want an untrusted network managing my AI devices, right? That's a huge risk. That's a risk as an enterprise. That's a risk as a public provider, a service provider. It's also a risk for a government. So trust is going to become increasingly important.
24:18Is there a timeline the industry is talking about in terms of 6G? Yeah, there's lots of different discussions. I think there's a view of end of the decade. I think we're going to see 6G services faster than that. Obviously, if you look at 5G, there's places where we don't have full 5G deployment. So I think what you'll see is this will play out over time. But my expectation is 6G will be the beginning of this transition to services that we deliver versus just these big forklift events that we've been used to in the past. Let's just get back to the data center discussion as well, because it's obviously central to all this AI discussion.
24:53There have been these big announcements regarding huge data center buildouts, a certain amount of gigawatts here and a certain amount of gigawatts here as well. But I'm hearing a bit more of this discussion around potentially smaller data centers, which are connected with possibly your technology as well. Often a term you scale out, right? You know, the connection of many different data centers to be able to deliver the capacity. Is that a trend that's happening? And if so, why? Yeah, I think a couple of things. First of all, you've got this concept of scale across. Scale across is the connection of multiple data centers to scale training.
25:32That solves the problem of some of the things you talked about, just the limitations of I can only get so much power here. I have this much space in terms of what I can put in. What we're seeing is as the demand continues, and especially as we get to more inferencing, right, a lot of the scale across problem was originally around solving training. As we move to inferencing, you don't necessarily need these large clusters. Now, those large clusters are still an asset because if you're not using them for training, great resource for inferencing. But you can build these smaller inferencing centers.
26:02And this is also important when you think about things like sovereignty and security and locality and latency even. So these smaller data centers become powerful. And that's where this data center interconnect with our optical and our IP technology. We think there's going to be just a massive boom. And I I think we also got to recognize, you know, even with some of the, you know, there's some pushback around, you know, there's been some pushback around data center build outs. I think that's, you know, I think those things, you know, those things will play out. But even with that, there's also a tremendous amount of data center infrastructure in the country that can be upgraded to support, you know, AI native workloads, right?
26:38Legacy infrastructure that can be modernized. And so when you think about that in the US or you think about it here in the UK, for example, or across Europe, that all is an opportunity as well to put modern GPUs or AI native infrastructure into. But that also needs connectivity infrastructure. And that's one of the reasons, particularly here in Europe, I'm so vocal about the need for digital infrastructure investment and connectivity investment. Because without the connectivity, those data centers become islands. And that won't work for AI. Right. Can networking, is networking good enough that, you know, data centers hundreds of kilometers apart, for example, can effectively operate like one unit?
27:17Yeah. I mean, with what we're doing in AI today, we see it in a thousand kilometers of distance. I mean, it's pretty powerful, the innovation that we're driving. And under the sea? Under the seas, that's probably a little bit too hard on the latency side. Okay. But at the moment, there are data centers connected under the sea, but these are more traditional legacy. Yeah, this would be more of a traditional connection. But the importance of that subsea connection for passing large amounts of data is obviously going to continue. Because I may train a model somewhere. I've got to distribute that for inferencing into different localities.
27:50So these long haul and subsea networks continue to be important. There were lots of interesting stuff you said in the last 30 seconds. And I want to pick up on all of Empower, Europe, and also some of the concerns around these data centers. First, Power. Everyone talks about the fact that these data centers can consume a lot of electricity and resource, etc. And often when we're talking about power efficiency in a data center, we're often talking about it on the chip level, for example, a GPU. But is there a chance or cooling or some other area, is there a chance to kind of gain efficiencies when it comes to power through the networking portion of a data center?
28:28Yeah, absolutely. In fact, this is, you know, we, we, I'm glad you brought this up, Arjun, because this is something with optical that's really driving the transformation. And there's a, there's a couple of things with physics as you get these faster and faster speeds. It means you have to power the copper more and more. And what happens is, is copper actually can't, it can carry really fast speeds, but the distance it can carry it becomes shorter and shorter. And so what do I have to do? Well, I have to then reamplify it. And that requires more power. With optical, I can carry much faster speeds over much longer distances.
29:02And I don't need to repower it. So it's much more power efficient. And so as we get into these faster and faster speeds, across data centers, but even inside the data center, we see demand emerging with optical. Can you just lay out kind of the difference between optical and copper? Yeah, really simple. I mean, copper is electricity delivered over a wire. and optical is non-visible light delivered over fiber. It's amazing, isn't it? Yeah. That's incredible. One of the areas you've been investing heavily in as well is sort of a lot more of your kind of stack, I guess, lasers. And I don't want to get too technical, but I think it's really important.
29:41There was a recent note from analysts at Rosenblatt and they said laser making is the moat at Nokia. And they said it's being built now. And I know you're investing in kind of indium phosphide and your own semiconductors and basically all of these components that are required for the optical connectivity portion of your business as well. Just what are the importance of lasers and chips in the products you're putting to market? Yeah, today, you know, all of these optical fibers need a laser to create the light. And so we connect them with, you know, with small computers. We call them digital signal processors.
30:16You may have heard that term before. So those DSPs actually provide the communication. Then they take the electrical signal and it provides the capability for us to convert it into a light signal. But I still need a light source, which is the laser. And what we do that's unique with our Indian phosphide technology is we build what we call an integrated circuit. So it's a very small chip that does all of the light service. So the light source, the modulation, the attenuation, even the receiving of the light on the other side. And the power, the benefit of that is that it's far more efficient in building these systems.
30:54As you go forward and you think about bringing optical inside the data center, you still need all those same components. But it's just, it's a different bandwidth, it's a different distance than what we might build for a subsea solution. And if you look at what we announced in March of this year at a big optical conference called OFC in Los Angeles, the announcement was oriented around building different products and platforms so we can deliver the light sources and the digital signal processors to make sure we're building light for a small, coherent application inside a data center all the way out to something we'll put under the Atlantic or Pacific Ocean.
31:34Right. And you're investing in capacity around lasers, around your own semiconductors for this solution, right? Yeah. And actually, in my Q2 earnings, we announced a partnership with an acquisition of NXP's fab in partnership with them to transfer their facility in Arizona to build more capacity for optical. Because this trend is not going to slow down, right? As speeds go up, we have to have more and more optical. So we know that we're playing into a long-term macro trend. So would you agree is laser making the moat at Nokia? Well, I think it's a moat. I like having lots of moats, but we're excited about our optical components business.
32:14Another interesting point you mentioned just a few moments ago was about Europe. And you said you've been very vocal about the need for Europe to build out its connectivity. There is obviously this large debate and geopolitical race to some extent around who's going to win AI or who's going to be a more dominant player in AI. And often the conversation centers around the US and China. And there is a view that Europe is behind on multiple areas, whether it's semiconductors, whether it's the models themselves, whether it's the ability to build the kind of data centers and the connectivity that the US and China have.
32:50What is your view here? What does Europe need to do to kind of be in that conversation? Well, look, I've been pretty vocal on this on a couple levels, but let me take an example of doing this right in Europe. And I'm going to pick – I'm probably a little bit biased because we're based in Finland. So I'm going to pick Finland. Finland has a great history of technology companies, not just Nokia, but you've got a quantum company, IQM, there that just recently went public. You've also got internet gaming companies. If you look kind of across the cycle, Finland's got a great group of technology companies.
33:24It really outpunches its weight, you know, given its size. So it's really a great player. What Finland has done, which I really applaud the government there, is they've worked out a deal with Google to have Google invest into Finland and invest and build data centers. It's creating jobs. It's underwriting the investment in power generation modernization, nuclear power plant generation modernization. It's also creating a bed for innovation. The advantage we have in Finland is that we have great networks. We have some of the world's best 5G networks, the best edge network and routing. We have some great customers in Finland and the telco operators.
34:06They're great examples of what is being done right. That's the kind of bed you need to build because the reality, if you think about it, you're an AI developer. You're trying to build an app or a service or maybe even a model. Well, where are you going to go to get that infrastructure? And today that answer is, I don't have anywhere to go in Europe. So I'm a big advocate of these kinds of partnerships. And I think it's great to, you know, sovereignty is obviously essential. The reality, and I've been quite vocal about this, the reality is to be sovereign, you also need to be competitive. And the reality is no one today is sovereign or truly independent.
34:41No one has complete independence from the rest of the world. There's no isolation. So sovereignty is about control, innovation, scale, and speed. And that's what we need to see, I think, more broadly in Europe. But I think Finland is doing it right. By the way, I think credit is deserved not just in Finland but across the Nordics. You see companies out of Sweden, lovable example, great example as well. This is the kind of approach and investment we need. And obviously there's some benefits just in terms of where Finland is located. We can leverage the water that's used for cooling, for thermal reuse, for heating.
35:23We have some advantages around green power because of our infrastructure. There's some things that Finland has structurally around its grid that are advantageous. But it doesn't mean that everybody has to solve it the way Finland does. But the mindset has to be, for us to be sovereign, we have to be competitive. And I think it's actually a pretty existential issue for Europe because the reality is the internet was largely about media. content, entertainment, consumer digitization. Obviously, there were a lot of commercial value as well. But if you think about those big companies, social media companies, search, Europe missed a lot of that.
36:02You can't point to a lot of players that were winners in the internet super cycle, right? That didn't happen. But it's more existential because think about the industries that are going to be transformed by physical AI. Vehicles, robotics, industrial automation. And I really applaud companies like ASML. I mean, what Christoph is doing, championing this push in Europe. I mean, they're an industrial automation company. They need AI to be competitive because he realizes that what he's building, which is leading edge semiconductor manufacturing equipment, he's not going to stay on the leading edge if he's not innovating at pace.
36:38And so he's leveraging that. You know, my push to Europe is, you know, let's use this as an opportunity to make big investments in digital infrastructure. We don't need to necessarily build frontier models, leverage, you know, great work that, like what Mistral is doing, what Arthur and team are doing around innovating in open models. You know, it's another great place where Europe can lead, but also drive the sense of urgency around innovation in its core industries. Because the alternative is it will get done by someone, right? And ultimately, what we miss sometimes in technology is technology is very much about speed and scale.
37:16Because what I don't get in technology, and Nokia, this is a company that's living history of it, right? You win one cycle. You're not given the right to necessarily win the next. And you obviously saw that with phones. So in technology, if you don't win the cycle, you may not be there for the next one. And I think that's the urgency that we have and why we're, you know, I'm really pleased that at least in Finland and to see some signals across the Nordics that that's well understood. It's also why, you know, we're trying to do our part. We're leaning in on the AI Gigafactory effort in Finland and making, you know, being the leader, are making commitments around that because we want to see a strong and competitive Europe.
37:59By the way, competition is good. It's great for the US. We'll see more innovation. As an American, I know that competition will drive more innovation in the US as well. And that means ultimately that the Western world will be competitive. And I think that's really important for the future of the Western world. Yeah. And just to clarify, Justin, when you say it's existential, it's because Europe's industry is built on manufacturing and so much of these traditional industries where if it doesn't modernize with AI, then there is that existential threat, whether it comes from US companies or Chinese companies or elsewhere, that if they're not modernizing with AI, then that's where it's existential.
38:35Yeah, that's right. I mean, you think about it, there's a small number of hyperscalers. They're all basically in China and the US. They were born of the internet super cycle. Many of them are investing aggressively in the super cycle. But the next step is, where's autonomous vehicles coming from? Today, it's largely the US and China. But who are the automotive leaders? A lot of the automotive leaders are in Europe. And if they don't invest to win, it's not... Regulating against it, of course, I'm all for making sure we have safety and security. These are really important. But you can't slow down the progress of technology.
39:14Do you think European regulation right now is hampering Europe's ability to compete in AI? I think on multiple layers. I mean, one is, I actually think Europe has done some really good things in regulation. If you look at the internet, it took quite a bit of time, but they came out with a very thoughtful regulation on GDPR. It's been good. It's been a model. In AI, the approach was let's regulate before it's matured. And I think that's actually a mistake. I think, again, fundamentals on safety and security, responsibility, very important. But regulating to the point that you slow down innovation is a challenge.
39:50The other thing Europe needs to do is allow and support and enable consolidation. I think the capital markets union, very important because ultimately this is a scale game. You support investment. You need to support risk-taking investment. So some of these other elements. And look, I want to acknowledge and give some credit to President von der Leyen because I think even though it didn't hit the headlines, if you look underneath what she said yesterday in her State of the Union speech, she touched on some of these things as being important, that Europe needs to lead in these spaces, it needs to invest.
40:25And she's pushing and challenging the European states to support these types of initiatives, support a capital markets union. So I'm optimistic that we can get there, but I think there has to be a much greater sense of urgency Because I think the risk of the current course is that Europe just won't be a competitor or player at scale in the AI super cycle.
40:52The AI safety debate has burst back into the public view. It began with the resignation of a former anthropic researcher, Jacob Coxon. He warned that the industry was gambling with potentially catastrophic risks. Then Evan Hubinger, who is another anthropic researcher, put an even starker number on it. He said he believes that there is a greater than 10 % chance that AI could kill all humans within the next decade. So then anthropic CEO Dario Amadei came out with an essay in which he called to pace the frontier. In other words, not to stop AI development, but to slow the race for ever more powerful models enough for safety testing, oversight and what they call alignment research to catch up.
41:34Amoday argues the technology could be hugely beneficial, but that the tech is moving too quickly before we understand how to control it. And he's actually found some unusual support, including from Elon Musk and OpenAI's Sam Altman. Now, these are all rivals. So it's very rare for these three to agree on anything, really, given they have competing companies. The big question now is whether the companies building this technology will actually accept limits on the race or whether competition and the fear of falling behind China, which is a very key point, will keep pushing them faster.
42:11You mentioned regulating around safety, and safety is a massive discussion as we talk around AI. And a lot of the discussion on safety has been led by warnings from some of the big US AI labs. So, you know, we had this resignation of an anthropic researcher, and there were claims about sort of AI's threat to humanity as well. You had the Anthropix CEO, Daria Amadea, and some of the other big Frontier Lab CEOs like Sam Altman and Elon Musk also backing the calls for a slowing down of the pace of advancement of AI as well. Do you share any of those concerns that have been raised by those model companies?
42:48Well, I think, first of all, responsibility is important. I think how you do it, and you touched on three that kind of advocated one approach. And Mark Zuckerberg and Jensen have stated a different view on that. But I think everybody has a view on doing this the right way. But I also think we have to balance, you know, we need to be responsible. Look, I'm not, I don't spend my time at the edge of the frontier of models. They're far more expert than I am. What I'm focused on is making sure that we build the trust and the security in networks because ultimately, you know, what we're going to be asking networks to deliver and how they deliver it becomes even more essential.
43:27And we saw, you know, we saw signs of this with, you know, with some of the attacks on, you know, cybersecurity attacks on networks in the past. So our focus is there. But, I mean, ultimately, you know, we need safe, responsible, you know, AI. But I also think it's a bit odd to me to think that a company is going to go in and say, I'm so focused on being at the frontier, you know, Western company, that I'm going to do something damaging because the impact of the company is so massive, right? It's destruction. I mean, you're basically going to damage the company. You're going to slow the industry down.
44:04So I think it is a really important part that they take responsibility. And for all of us in this ecosystem, making sure we're building the security, the safeguards, the right architectures into our layers of the stack. And that's where we're focused. If there is any kind of agreement on slowing down the pace of development, are you concerned about any impact in terms of the demand for the infrastructure build out right now? Yeah, well, look, I think the reality is that even where we are today is so early in the deployment and the power of these models is so massive that there's a lot of demand to run and just deploy.
44:41I even look at us internally. I mean, I'm really pleased with the progress we're making. And we can go and be really aggressive in marketing what we're doing. But if I look at where we're making progress, material progress, it's in a few places. There's so many more things we could go do inside the company, which, by the way, include things like, you know, helping provide the security and the trust in these networks. And I think that's true when you look across the market. And by the way, that's just even if we didn't have another frontier model released in the next three years, we could probably make tremendous progress just deploying the technology that's there today.
45:17So my sense is there's actually quite a bit of runway either way. but I also think we're going to find a way to continue to accelerate progress in the industry in a responsible way, which is, by the way, what has been happening the last couple of years. And another sort of issue linked to this is some of the opposition we've seen, I think, particularly in the US to the building of data centers with residents in certain communities worried about electricity prices, around grid capacity, water, land use, all of these various issues as well.
45:48What's your read on this? Is this something that could derail some of the efforts around these data centers? Does the industry need to be more of a working with those communities? What's your read on what's happening? Well, I think this gets, unfortunately, there's probably some bad examples that create a lot of news. And then I think it becomes a political issue. I think when you look at the facts, there's a lot of companies doing this responsibly. And when done right, when it's done right, and I talk to a lot of these customers about this, and I see examples of this, obviously, it goes back to even my time building a supercomputer in Finland.
46:26When done right, it actually lowers the power cost for the community. It brings benefits to the community in terms of tax income, which helps the community make investments. It creates jobs in the community. And ultimately, it provides a structure for investment and economic development that can benefit the community. So I think, do we need to be responsible in building these? Absolutely. Should we be so binary that it's good or bad? I think that's quite a simplistic argument and not the right one. And I think there's actually many more examples of this being done right. Just like if I look at what Google announced in Finland last week, it's a phenomenal example of data center investment done right.
47:11Let's spend the last part of this conversation, Justin, just talking about something a little bit more futuristic, but I think interesting nonetheless, space. Modulate. You've been asking me about space. I've been asking about space. Yeah. I mean, when we spoke, I think, in February this year at Mobile World Congress, which is usually where we meet, I asked you about space. And I think the conversation's got a lot more interesting, I think, since the IPO of SpaceX, of course, with Elon Musk. And, you know, one of the points in there and his goals is to effectively build data centers in space. And actually, he's not alone.
47:46I mean, we've seen Google looking into this kind of project and others, too, as well. But first, I just wanted to ask about Modulate, which is a space communication solutions business you've got that was spun out Nokia Bell Labs or created within Nokia Bell Labs. What is the purpose of this? And I know you're also the first company, I think, to create a telecommunications network on the moon. All of these projects, what is the point? Yeah. Look, these are a couple of things. One is, you know, Bell Labs is a phenomenal institution. And actually, we don't talk about it enough. But if you go look at AI RAN, Bell Labs is behind AI RAN.
48:23Bell Labs is also doing things around optical innovation, next generation fiber networks. I'll be able to, if you're worried about your bandwidth at home margin, the good news is in a few years, the Bell Labs engineers will make sure you can have probably eight to 10x the speeds you have today because of the kinds of innovation. Thank goodness, because there's a lot of good games coming out. They require a lot of bandwidth. So, look, we're hard at work at that. And it's an institution that you constantly, if you get back to its core, it constantly has to reshape and readjust itself because its core ethos, going back all the way to Alexander Graham Bell, was deliver groundbreaking innovation that has commercial application.
49:04The reality is some of that groundbreaking innovation doesn't have commercial application that we can take care of. But we've been doing a lot of work and space. And Modulate was a great example of a company that for us wasn't a fit with where we're headed, but it has technology value. So we spun it out. There's other work we're doing around space communications, anticipating data centers in space, but also thinking about how to help our ecosystem, our industry around connectivity, add the layer of non-terrestrial communications that those researchers are working on as well, and that will yield new value.
49:42And so I'm really excited about where we're headed there. I think non-terrestrial communications, data centers in space, and the connectivity layer of that, those are all going to be things for us that are new opportunities. And a great example of this, and there's a lot of innovation going on in this space, but we talked about those lasers. I mean, it turns out the way you're going to get communications from, you know, from space down to the Earth is not big cables, right? It's going to be light, but it's still lasers. So it's, you know, it's still delivering, you know, light-based communications at really high bandwidth is one of the ways we can do it.
50:19And so a lot of the innovation we have gives us, you know, great potential to leverage that. But that's why we need that innovation with Bell Labs. That's really interesting. So just more broadly, why is everyone exploring space as an option of either building these kind of data centers or cellular connectivity and other things? Yeah, so it provides a cost-effective way. It's an additional way to provide capacity. I mean, it's an obvious example. If you think about what we're talking about in just mobile or home connections, you can provide connectivity in places where you can't provide it otherwise.
50:55Otherwise, you can provide additional capacity where you may have gaps, right? Or there's an issue with the network. You can provide critical communications. It's complementary to what we do. Just like when I think about it, 5G and Wi-Fi are complementary technologies. And specifically for data centers in space, because it's something Elon Musk has said he wants to do, is it possible technically and from a cost perspective as well? And is it something you think we're going to see? Yeah. So one, it's absolutely possible from a technical perspective. Two, I think around the cost is really, you know, that's a question probably for some of those players.
51:33But I think if you look at it, the cost of, you know, like all of these things, as you drive the innovation agenda, cost comes down. And I think the other thing you have to look at there is look at what Elon has done just disruptively to his credit on putting rockets into space, right? I mean, no one thought you could launch a rocket at the kind of cost he's doing. So as you bring that cost down, which is the biggest cost of getting something into space is the rocket itself, right? So as that comes down, then the economics of putting a data center up there become more and more attractive. From our part, we want to be ready to make sure that where we can be relevant in that solution, we're delivering the innovation and the technology needed to make sure it can happen.
52:11Have you sized up the opportunity at Nokia of space and data centers and connectivity? Yeah, we think it could be massive. Yeah. Do you have a number? No. Not that I'm giving you. Justin, just finally, in the next year or two, what would you say are going to be Nokia's biggest growth drivers? Yeah, look, I think today, we look ahead, it's absolutely going to be driving the demand in data center infrastructure, whether optical and IP networking. And then I think what we'll see, which you won't see in the numbers as much as the early penetration of AI ran as we start to see physical AI build out.
52:48And just finally, and this is probably the question everyone's most interested in, will Nokia ever return to phones? I genuinely think if Nokia returned to phones, there would be a massive audience who'd want them. Well, right now we're hard at work capitalizing on all the opportunities we have. But Arjun, we've got a great, one of the things we didn't talk about is we have a phenomenal licensing team that enables technology everywhere. So if you want to build a business with the Nokia phone brand, there's a company doing it today. But we're always open to a conversation. Justin, it's been a pleasure.
53:22Thank you so much for your time. All right. Thank you, Arjun. This has been great. if you'd like to get in touch you can email me on the tech download at cnbc.com or i'm on tiktok instagram linkedin and x you know all those internet places reach out let me know what you thought of the conversation that's it for this episode of the tech download thanks for listening and watching and i'll catch you next time
53:55Thank you.
Nokia CEO Justin Hotard joins CNBC’s Arjun Kharpal to explain why the company sees itself at the centre of the AI infrastructure buildout: providing the optical and IP networking technology linking AI data centres, as well as the mobile-network equipment that could support the next generation of robotics, autonomous vehicles and other physical AI.
Hotard argues demand for AI infrastructure remains in its early stages, with customers constrained by power, silicon and supply rather than a lack of appetite to build. He also lays out his view of 6G: less a conventional upgrade in mobile speeds, and more a shift toward intelligent networks that can sense, plan and act.
The conversation also covers Europe’s AI challenge, the debate over data-centre expansion and Nokia’s ambitions in space communications and orbital data centres.
See Privacy Policy at https://art19.com/privacy and California Privacy Notice at https://art19.com/privacy#do-not-sell-my-info.