Nvidia’s CEO on the AI debt scare, China, and jobs | Full interview

Hi everyone, I'm Dear Derbosa live in Berlin and I am pleased to be joined by Nvidia CEO Jensen Huang fresh off the stage here at GTC Berlin. Thank you for coming right to me. I appreciate that. >> Delighted to be here. >> How's the Europe tour? >> Thank you. >> Incredible. Incredible journey. We started in uh I guess we started in France or London London to France to Sweden to here and uh last in the in the last year we we've invested about 10 billion dollars in uh companies here and uh this week we announced 14 more. So we're in the process of building the AI economy here. >> Right. And some of your message has been just start building, start going overthinking, stop overthinking, start building. Do you think that message got through? >> I hope so. And you know the chancellor chancellor Merz also told everybody to lean in. Let's let's go build this thing. You know I think the the fact of the matter is is the technology had to be invented for the last 15 years where we're kind of in invention mode. But this six months ago you know we really transitioned to AI being useful mode. And the thing that I wanted to do during keynote was to demystify >> um what makes AI useful, what what are the components of AI and why they could build it, you know, incredibly here, >> right? And you talked a lot about that on stage just moments ago. I want to talk about one of your announcements, DSX pod, an AI factory in a shipping container. Is that a fair sort of representation of it? You know, it's it's fascinating because it gets at the supply chain, but it also gets at this financing question that's been sort of rattling the AI trade over the last few weeks, especially the Neoclads. So, if you can give lenders essentially more certainty about what a data center costs and when it starts earning money when it comes online, >> do you think that makes it easier to finance? >> This is going to make it way easier. Let me tell you why. So, a gigawatt of data center is like $50 billion. And that $50 billion from the moment that we ship the technology components, it goes through an OEM, those computers land at a data center. From that point to operationalizing could be four to six months. And during that four to six months, you're wiring these computers up. You're getting them all booted and and so what if we were to do some of the manufacturing of that in the supply chain? The advantage is that you might be able to shorten that four to six months to maybe three to four months or take three to four months out of it and uh as a result of that your time to revenues is going to be accelerated but also that idle cash because you had to pay for it that idle cash would be reduced >> right so it's sort of narrowing the cone of uncertainty right and that has been the major concern right now markets um are sort of they're volatile we'll say worried about neoclouds then optimistic um what do you think the chances of a shakeout out any kind of shakeout are >> I think this is going to get solved and the reason for that is because the demand is so good you know just last night SpaceX just announced and I think Elon uh said that demand is many times >> the capacity he has and he has a lot of capacity and so he's building more because the amount of demand is just incredible coreweave is kind of the same way uh NBS is the same way and scale is the same way I think the challenge of course is that um and this is kind of what you were trying to get at from the moment that the computers are shipped to the moment that they start earning um could be many months and the amount of float because we're talking about a fair fairly significant uh investment the amount of float is quite large and so uh we need to help we need to help these companies uh reduce that and I think this the ex reducing the supply chain making it easier for them to stand up these the the computing is going to make a big difference >> right so you're doing this on the Nvidia side but what about some of the physical bottlenecks like power how much of a timeline can you really take out of the equation if the customer, you know, is waiting years for grid connection. >> This is really one of the reasons why we're so resilient to this because we have so many different customers. You know, we're the only we're the only computing company that supports every single AI model, every single CSP, every single computer company um across NeoClouds in every single country. And so, uh, there's some there might be some bottlenecks somewhere, but because the overall demand is so great, um, we'll be able to get the computing out to the to the market wherever they need it as soon as possible. >> Right. And what is your outlook on the power question though? Some of those physical bottlenecks. Do you think that gets solved in the next few years? What, you know, there's a lot of efficiencies on the chip side on what you're doing with the DSX pods. Um, but what is sort of the bigger picture there? And do you think that's the biggest bottleneck? >> Um, it is one of the major bottlenecks. uh the bottlenecks keeps shifting and the reason it started with with wafers and to co-wash to HBM memories to uh silicon photonics to even connectors and um as those bottlenecks are they're going to continue for probably a couple more years. Uh now that we're ramping up these AI factories around the world now power becomes uh increasingly so. But you know oftentimes the power is available but it's the land power and shell is either maybe a construction limitation or maybe the power generator is on being delivered or connection to the grid or you know whatever you know a lot of construction challenges because the the number of jobs that this has created for construction workers around the world is incredible and so uh you know there's there's limitations there and so anyhow I think the bottlenecks just uh uh these are physical bottlenecks and they're part of the physical world. >> Are you are you sort of getting at some of the permitting bottlenecks as well getting these online? >> Some countries have permitting issues but some companies some countries don't. Some states do but some states don't. And so I think the the NVIDIA diversity uh really makes us quite resilient to this. uh you know, we'll be able to ship it somewhere and whoever has a bottleneck um we'll work we'll work with them to to make sure that as soon as they get the bottleneck resolved, we'll ship computers to them as fast as possible. Meanwhile, we'll ship the computers to somebody who can use it. >> Right. And you're standing ready. Um I want to talk to you a little bit about compute as a commodity. It's starting to trade like one. There's GPU futures on the CME. Nebas has moved to spot pricing that can change every 15 minutes. Still early days. What does it mean for Nvidia when computing power gets treated like capacity? >> Well, this is one of our advantages because Nvidia's computing systems are so universally used and the performance is is so good that's productive for so long and because we're software programmable and we keep enhancing the software throughout the entire life of the the asset. We keep refreshing it so that it's productive for a very long time. And so people are discovering that what people thought used to be a four-year life cycle really is probably eight to maybe more. Um and irrespective of of um uh of uh uh the clients that the customers that that use the compute, they're always companies that can use the compute and so there's just so many offtakers. So it's productive, it's fungeable, and it's durable. Now when that happens you might be able to uh underwrite a loan against the asset itself. Today most computers you know don't have value don't have very little residual value but people are now recognizing Nvidia's compute has a great deal of res residual value and so you might be able to use that and you know backs stop a the debt itself. are some of the financial instruments and derivatives around the AI space. Does anything about that worry you? >> Well, I think that that one of the pillars for this growth because we're talking about infrastructure buildout, the capital market will have to uh think about how to support it and now there'll be different capital instruments. Um but I think the the real advantage that we have is because of the durability because of the productivity and fungeability of our computing system the risk involved in underwriting NVIDIA's asset in our NVIDIA computing is the lowest of any asset out there and so I think this is an area that's going to really play into our advantage. >> Right. Um you mentioned the durability. Um, one of the most like surprising things of the last few years has actually been the durability, the lifetime of chips, even older generations. There's some really great data from silicon data that shows, you know, Blackwell rental prices are still rising. Um, I think you mentioned durability. Did you say seven, eight years a moment ago? Is that what you think the durability is for a Blackwell? >> Probably. >> Um, >> what do you think that is for a Vera Rubin, the next generation? Well, I think that that um Vera Rubin is probably going to be the same. It'll be, you know, seven, eight years and and um of course the technology is changing so long as so long as uh we continue to upgrade the software. This is where Nvidia's CUDA is such a big advantage. We're introducing new algorithms, new software layers all the time. And when when we invest in uh enhancing the install base, you'll be able to continue to use it for a long time. >> Um let me bring in Gavin Baker. He's waiting in the wings. He was my GTC pregame host. >> Is that right? >> Apparently, you guys did pretty amazing. Superstars. >> We We had a lot of fun. We talked about safety. We talked about open- source models. Um, one topic that we didn't get to a huge amount, which I've sort of been talking to you about now, is sort of >> Gavin, >> how are you, my friend? Good to see you, brother. >> You know, this is why we do this live because we can have surprises here, right? So, Gavin, thank you for for joining us. Um, you know, I've been talking to Jensen a little bit about some of the jitters in the markets, which, you know, always come around. I wonder how seriously do you take these ones now. And are we sort of out of it? Comments from Elon Musk, as you mentioned, on the earnings call. We've got Jensen here right now saying that demand is through the roof. Is there still that financing risk that's being overestimated or underestimated? Well, when I think about financing risk, I read something incredible last night, an open- source model, DeepSeek, said that they are monetizing at $38 per GPU hour. Last time I checked, like I haven't heard anybody renting GPUs anywhere near that high. So maybe this explains Blackwell prices up 17 days in a row. >> Yeah. How do you think about the intersection what is clearly extraordinary demand >> with supply from a capital perspective? >> Well, you know, the single the single problem that nothing can solve is demand. The fact that there's so much demand across the world. I mean, this isn't just from one model. We're talking about open open AI has got incredible demand. Anthroy has incredible demand. Metamuse has got incredible demand. You've got Grockbot, incredible demand. Uh Gemini for just coming out. Uh you just mentioned Deep Seek. Uh Moonshots models got great demand. GLM's models got great demand. And so you got all these different models, closed models, open models. The demand is just extraordinary. And we're just we're going into this. We're really actually going into the inflection point of useful AI um as we speak. And people are more people are using it. It's they're using it to do more useful things. And that's the reason why token demand is so high. I think all of the other stuff is they're going to get resolved. You know, without demand, nothing is resolvable. >> Absolutely. >> Yeah. With demand, everything will be resolvable. >> Demand can be incredibly high and usefulness can be going up, but if the financing is illtimed and you've heard people like Daru Ammoday, Satia Nadella, others talk about this um that can sort of trump everything else. How much are you thinking about sort of the macro environment? We have to make it uh easier for people to finance these buildouts. And one of the areas that that um uh we have a real advantage in is what you and I were just talking about the fungeability and the durability of the asset. You could use the asset to underwrite the uh the loan. And so the fact of the matter is you know the residual value of Nvidia's compute is really really high >> at the moment. Nobody gives a credit for that. I do think we should, you guys touched on it a little bit earlier, but the DSX pot I actually think is pretty important here. Like if we think let's use a trillion dollars as a nice round number of AI capex seems like we're going to go much higher very soon. >> But a trillion dollars if it's a 6 to9 month cycle time. >> That's a lot of wasted capital. That's right. That's unproductive. So maybe talk about the DSX pod and how that changes. >> The moment it gets shipped, somebody has to pay for it. But the moment then you have to install these AI factories and they're gigantic and putting these AI factories into service takes months. If we could reduce that months, you know, from the moment that the systems are shipped to the moment that it's deployed and and being operational, if we could reduce it from months to weeks or even a month, that will be completely gamechanging for all of these AI factories, right? Having a trillion dollars, having a few hundred billion dollars sitting idle while you're waiting for revenues is obviously a challenge >> and can give your customers a little bit more control over that sort of very pressing, very important financing question. Gavin mentioned open models. Um, we've had sort of a year where it's just exploded in terms of adoption in the US. It's been a lot of Chinese open models in the spotlight, but over recent weeks, we've got some really interesting big developments from Western companies. Reflection, Mistral, Meta, of course, there's Neotron that you talked about in the keynote. Um, how do you see the balance right now in open >> openw weight AI? >> The world's going to use closed models and open models. Uh, in the last 6 months of this year, the first six months of this year, several hundred billion dollars, I think $600 billion worth of venture financing went into AI native startups. 80% of those AI native startups are using open models. And so, uh, you're able to take what is close to a frontier open model, um, post-train it with your expertise and your data, turn it into essentially a super AI, >> um, that is good at your particular domain of use, and build a company around it. And so you're starting to see this all over the place, whether it's in in, you know, legal services or uh entertainment or um uh manufacturing and design and all all these different applications are now jumping on. And so so I think I think the world is going to have both closed models and open models. You're just going to use the best model where it makes sense. >> And we heard from I think every single guest in the pregame that they're using a mixture. They're using a number of models. Um so far it's been the Chinese ones that have seen a lot of adoption. Do western models catch up? Do we see sort of them gain momentum given, you know, the releases of the last few weeks? >> Yeah, no doubt. I mean, um, Neimotron with post- training gives you near frontier. You know, when when you're trying to build something that's domain specific and application specific, you don't need to have the frontier of general intelligence. You just have something that's close to something like reflection or TMLS is pretty great. Nvidia is great. And you take these models, you post train it with your own data. When you do that, they become super intelligent at the application space that you want to target. And so this is a this is, you know, I think the the open models has reached the frontier level. And I think at this point we're going to see a lot of super intelligence built. >> Where does that leave the frontier labs? they're still going to be offthe-shelf incredible and and um I would recommend that everybody rent an closed a frontier lab closed AI model whenever you can and you build your own AI only when you must and so for a company uh it could be uh uh one of the one of the companies here in Europe do they have to build their own AI cadence has to build their own AI seammens has to build their own AI they need sovereign that's right they need sovereign AI so they need the ability to build their own custom proprietary ary super agents and um all of the technology all the resources are now put together. Um everybody should be able to do it and so you know I would I would just continue to use use off your shelf whenever you can and build your own super agent whenever you must. >> Well it feels to me like what a lot of at least our portfolio companies are doing is they are combining their own specialized intelligence. They're generally taking actually an American open source model, post-training it, making it amazing at the tasks they use it for, and then you combine it behind a router with your favorite closed frontier model. That's right. >> So, I think the the future is hybrid. That's right. >> Behind a router. >> Yeah. Just as computing today is hybrid cloud, today in the future AI will be hybrid AI. >> Yeah, absolutely. >> I want to get back to sort of this the China question as well. um you know just saw a recent report that Deepseek and Huawei are working together on chip designs and you have long said that restrictions would actually accelerate China's own industry. How do you read this moment and do you think it's time to take China's chip ambitions more seriously in America? >> We should always take China seriously. If there's one country that could manufacture things at scale, whether it's uh manufacturing of course chips at scale, manufacturing open models at scale or even very importantly manufacturing AI researchers at scale. Uh nobody does it, you know, more seriously at a larger scale and and better than China does. And so we we've got to be, you know, we have to be re we need to recognize that that we have very very serious competition. Um, but you know, the United States, we're, you know, we've got a lot of things going for us and and, uh, our frontier labs are the world's best. Um, our technology, our chips, our computing systems, our infrastructure, all the world's best and, uh, you know, we're going to continue to make sure that we stay ahead, >> right? We should compete on innovation. Has that message been getting through to Washington, to lawmakers, especially as we approach sort of a midterm? >> Well, to some. I mean, you know, obviously the the whole point of politics is that there are different people with different thoughts and and um different beliefs. U but I I think that at the moment um I would I would I would say that that um the US government would encourage us to go compete around the world. Uh there's a reason why we're in Africa. There's a reason why we're in uh Central Asia. Um uh we would like to compete in Brazil. We like to compete in Egypt and uh different parts of the Africa. We want to compete everywhere. And so so I I think um I the the policies the environment um makes it possible for United States American companies to to be the infrastructure for uh the European European countries um uh the global south uh so on so forth and I think we ought to go compete for it. One more question sort of in how this time how AI is perceived in the west versus the east and in the US you know there's a lot of push back on AI when it's mentioned in a graduation speech of course there's you know the data center permanent push back etc. China is a very different story and I you know they're embracing it going back to useful AI they find it useful why do you think it's developed that way >> well I think I think the the narratives around AI of course safety is important and we have to talk about safety and and the the most important thing about safety is to build the technologies so that we could deploy we could build these AIs and deploy the AI safely and so open shell is about that century is about that we've dedicated an enormous amount of R&D so that we could create these safety systems that the world could deploy uh safely and so so obviously safety is important. However, we also know um that this is really a incredibly powerful technology and and we need to make sure that we uh direct it in an optimistic way so that its potential and its positive impact could be enjoyed by society and and we ought to talk about that more. Um I think some of our narratives uh about about jobs uh obviously scare people. The facts are that uh AI has created more software jobs. The facts are that AI has created more healthcare jobs. The facts are even AI is creating legal jobs. Um if you look across the entire five layer cake that I talk about, the energy sector is booming. All the companies in the energy sector are booming. this is the best chance to invest in sustainable energy at no better time than you know now in the last 100 years >> and so whether you're a nuclear fusion or fusion or large or small nuclear reactors it's all being invested in right um battery technology solar technology it's all being invested in so so I think the energy sector is doing great obviously the chip sector is doing great the infrastructure sector is doing great um construction workers we're re-industrializing the United States we're re we're going to re-industrialize um a lot of Europe And so those manufacturing jobs are doing great. And so each one of these layers are creating new jobs. Uh so that the the narrative about jobs is almost precisely backwards. It's precisely wrong. And so so I think that that the facts are uh it creates jobs and I think that people are lean into it. >> I mean there's a narrative problem but there's also the way that sort of tech has developed and there's certain things. Microsoft, for example, took away the NDA, non-disclosure agreement, providing more transparency to these communities. Are there things that can be done, not just said, that would help sort of the reputation back? >> You bring you bring up a really good point about about um these AI data centers and AI factories. You know, when you go into a community and you want to build into local community, before you start building, you ought to extend yourself out to the local community and and partner with them. um let them know that what you're about to do uh is going to bring of course uh all the benefits but it it's going to of course bring new neighbors and we ought to be good good neighbors and I think that that uh the first couple of years and in building out the AI factories and data centers around the world um people were uh a bit too much of a hurry um they've everybody has now taken a step back whether it's the NDAs or uh contributing to the the communities whether whether it's uh building uh shops and centers ers and and community centers and improving schools and uh in many cases even even uh uh supporting lower taxes and you know all kinds of benefits that can be be brought to the local communities. I think those are all good ideas and uh each one of them ought to be ought to be uh tailored for the needs of the community. But the most important thing is you ought to be a good a good visitor. You know, you should be a good guest and before you go into a community, uh stop by and see if you could be helpful. >> A it's clearly creating a lot of jobs. It's doing amazing things for the world. We're re-industrializing America and Europe, helping to make America and Europe greener. >> Yeah. But to ordinary Americans and Europeans like how can how can AI make your life better today? What are a couple of tangible examples? >> Yeah, in a lot of ways uh almost everybody's using AI already and and um maybe people are looking for that extraordinary transformative phase shift um and all a sudden all of a sudden uh everything is different. Um I think everything is just going to be better as in technology over the years. For example, our search has gotten better. Um uh our ability to to uh use computers to write, to design, to create, to solve problems have have gotten better. I obviously obviously our our healthc care uh the ability to diagnose disease has gotten better. The number of examples of of uh discovering you know early detection of colon cancer or breast cancer, so many different examples, so many stories. um of of AI helping uh diagnostics that's gotten better. Um obviously we're going to make cars safer. Um without AI that's not possible. Uh we're going to make we're going to make planes more energy efficient, fuel efficient, not possible otherwise. and and so designs of new wings and new engines and um but I I think that the this the simplest idea is that AI is a tool that's going to amplify your capabilities and and um I I I don't know anybody who would want to have this incredibly useful tool to amplify their own abilities. Uh if you were a small small uh shop somewhere and uh uh you just you know a lot of things about the shop the business you're in but you don't know very much about software programming and you would like to create a website and this website you know is going to have this incredible digital marketing expert inside you could now get an AI to write that for you. So every single smallmedium business whether it's a a one of the one of the medium industrial companies Mostad uh here in in Germany uh if you don't have software capability now you have software capability and so uh everybody has all of this available to them and so I I think >> it amplifies capability but I think what many Americans are worried about is it amplifies risk as well and it's always that trade-off on even like a micro level you can see that with the rise of personal agents right? Like Muse, Grockbot, Instinct, and some others. Um, >> that's what Open Shell is created for. >> Exactly. Um, >> do you think that ultimately, you know, defenses can keep up with that risk? We talked about this in one of our panels. Um, but like when does that happen and what needs to happen for people to actually feel? >> We should apply the philosophy that that that uh is captured in OpenShell instantly. And basically it works like this. Uh AI is an auton these agents are autonomous systems and when you have an autonomous system on your computer autonomous agent on your computer you should first start with zero trust right >> and give it minimum privileges minimum access minimum minimum use of tools maybe they could access a file but they can't delete the file maybe can they can access an internet but they can't you know send stuff out and so there's a lot of different things that you could specify. We just need th those practices about how to deploy agents and deploy autonomous systems has to be adopted across the board right away. >> Um, and I think I've heard you use the analogy of like if you have a one-year-old, you put them in a play pen and you don't give them a hammer or anything that's going to hurt them. But the example that many people are now familiar with are agents or swarm of agents escaping this play pen or escaping that sandbox. Yeah. >> I wonder how much damage did that do to like the narrative to trust and was it a one-off or do you expect more of these things to happen? >> It it shouldn't have happened and it didn't have to. Yeah. It shouldn't have happened. It doesn't need to happen. If open shell was used uh it probably wouldn't have happened and hugging face uh uh developed an exercise a simulator and they came to the conclusion if open shell existed um their website wouldn't have been wouldn't have been attacked that way. And so I think the the the technology um now exists the methodology of of uh delivering software in a zero trust environment. Um that methodology has existed for some time. We ought to take advantage of that. And so I think we ought to build on existing technology. We had to create some new technology. Um and uh I the way to think about safety is we have to accelerate AI safety and and um safety technology and capability to me is one and the same. you know, how could you have a car that's not safe? And so, as much as it is capable, you want it to be safe. The two of them are really one and the same. And and so in the future, uh, as we de develop these AI capabilities, safety will come along with it. >> And this is very much your philosophy, Nvidia's philosophy. Do you trust the rest of the industry to have that same focus, focus on capability, but safety along the way? >> Yeah, I think so. I mean, look, Open Shell was announced. We have over 100 companies supporting it. Um I highlighted one particular collaboration which is what I'm very happy about which is anthropic uh using open shell working with Nvidia to create claude um agents that claude managed agents uh really terrific work. We're working with all the labs and working with all the operating system companies. I I think I think the the um incident uh unfortunate um but uh I I think that the it's it's a very very much has transformative for the industry >> and we have an existing system of laws and liabilities that incentivize companies to use the uh best play pins available >> and the play pin just got a lot better in the last month >> and it just it it history has not been kind to unsafe companies. No, safe and fast. No, there no customers and many of these customers are uh businesses other businesses and so these other companies can't put themselves at risk. >> So you're saying there's like a market reason to be safe and those that are not will drop out which I think is is fair. Um Jensen, thank you so much. We really appreciate you coming right off stage here and joining us live. Gavin, thank you for stopping by. >> Thank you. Thank you to both of you. Good times. >> Yeah. Fantastic. Thank you. >> Thank you.