NVIDIA’s Jensen Huang on Reasoning Models, Robotics, and Refuting the “AI Bubble” Narrative

Happy New Year coming up. Yep. Happy holidays.

Yeah.

and, you know, being in the middle of the vortex with it, what do you reflect on and say, like, this surprised you most, or this is the biggest change? Let's see. There there are some things that didn't surprise me.

Like, for example, the scaling laws didn't surprise me because we already knew about that. The technology advancement didn't surprise me. I was pleased with the improvements of grounding.

I was pleased with the improvements of reasoning. I was pleased with the connection of all of the models to search. I'm pleased that there are now routers that are in front of these models so that it could, depending on the confidence of the answers, go off and do necessary research and just generally improve the quality and the accuracy of answers.

I'm hugely proud of that. I think the whole industry addressed one of the biggest skeptical responses of AI, which is hallucination and generating gibberish and all of that stuff.

experts to actually be able to do what they do much better. That's right. And so in a lot of ways, I was expecting it, but I'm still pleased by it.

I'm proud of it. I'm proud of all of the industry's work in this area. I'm really pleased and probably a little bit surprised, in fact, that token generation rate for inference, especially reasoning tokens, are growing so fast.

Several exponentials at the same times, that seems. And I'm so pleased that these tokens are now profitable, that people are generating I heard somebody or it hurts heard today that OpenEvidence speaking to them, 90% gross margins. I mean, those are very profitable tokens.

Yeah. And so they're obviously doing very profitable work, very valuable work. Cursor, their margins are great.

Claud's margins are great. For the enterprise use of OpenAI, their margins are great. So, anyways, it's really terrific to see that that we're now generating tokens that are sufficiently good, so good in value that that people are willing to pay good money for.

And so I I think these are are really great grounding for the year. I mean, some of the things that the narrative that that of course, the conversation with China really, really, you know, occupied a lot of my my time this year. Geopolitics, the importance of technology in each one of the countries.

I spent more time traveling around the world this year than just by any time in the hit all of my life combined. You know? My average elevation this year is probably about 17,000 feet.

You know? And so so it's nice to be here on the ground with you guys. And so so I think geopolitics, the importance of AI to all the nations, all worth talking about later.

You know, of course, I spend a lot of time on expert control and and making sure that our strategy is nuanced and really grounded and promotes national security, but recognizing the importance of various various facets of national security. A lot of conversations around that. You know, of course of course, lots of conversation about jobs, the impact of AI, energy Mhmm.

on three points in space, three points in time. Now Mhmm. Maybe very near future, and then some some point out out in the distance, maybe some counternarratives, something else to think about with respect to jobs.

In the near term, one of the most important things is that AI is not just AI is software, but it's not prerecorded software, as you know. For example, Excel was written by several 100 engineers. They compiled it.

It's prerecorded, and then they distribute it as is for several years. In the case of AI, because it takes into the context, what you asked of it, what's happening in the world, right, contextual information, it generates every single token for the first time every time. Mhmm.

Which means every time you use the software in in everything that we do, AI is being generated for the first time ever, just like intelligence. Our conversation today relies on some, you know, ground truth and some knowledge, and but it's every single word is being generated for the first time here. The thing that's really, really quite unique about AI is that it needs these computers to generate these tokens every single time.

I call them AI factories because it's producing tokens that will be, you know, used all over the world. Now some people would say it's also part of infrastructure. The reason why it's infrastructure is because, obviously, it affects every single application.

It's used in every single company. It's used in every single industry. It's used in every single country.

Therefore, it's part infrastructure like energy and and Internet. Now because of that and the amount of computers that's necessary to generate these tokens, and it's never happened before, and because we need these factories, three new industries have emerged. Number one well, three new type of plants have to be created.

Number one, we have to build a lot more chip plants. Mhmm. TSMC is building right?

SK Hynix building a lot more plants. And so we need more chip plants. We need more computer plants.

These computers are very different. These are supercomputers that the world's never seen before. Right?

Grace Blackwell looks like a very different type of computer than anything that's ever been made, and entire rack is one GPU. Mhmm. And so we need new supercomputer plants, and then we need new AI factories.

Mhmm. These three plants are currently being meant being built in The United States at very large scale Mhmm. Quite broadly all over The United States for the very first time.

Mhmm. The number of construction workers, plumbers, electricians, technicians, network engineers. Mhmm.

You know? Right? The number of the skilled labor that's necessary to support this new industry in the near term, it'll be enormous.

Let's just face it. I'm so excited to hear that electricians are seeing their paychecks double. They're being they're being paid to travels.

Like like us, we go on business trips. They're going on business trips. And so it's really terrific to see, you know, that these three industries are now three types of plants, factories are just creating so much so much jobs.

The next part is the the near term impact of AI on jobs. And one of my favorites is, I love Jeff Hinton. He said, you know, some five, six, seven years ago that in five years' time, AI will completely revolutionize radiology, that every single radiology application will be powered by AI, and that radiologists will no longer be needed, and that he would advise this the first profession not to go into is radiology.

And he's absolutely right. A 100% of radiology applications are now AI powered. That's completely true.

And in some eight years' time, it is now completely pervaded radiology. However, what's interesting is that the number of radiologists increased. Mhmm.

And so now the question is why? And this is where the difference between task versus purpose of a job. A job has tasks and has purpose.

And in the case of a radiologist, the task is to study scans, but the purpose is to diagnose disease. And to do new research. And and that exactly.

And to do new research. And so in the case in their case, the fact that they're able to study more scans more deeply, they're able to request more scans, do a better job diagnosing disease, the hospital is more productive, they can have more patients, which allows them to make more money, which allows them to want to hire more radiologists. And so the question is, what is the purpose of the job versus what is the task that you do in your job?

And and as you know, I spend most of my day typing. That's my task. Uh-huh.

But my purpose is obviously not typing. Mhmm. And so the fact that somebody could use AI to automate a lot of my typing, and I really appreciate that, and it helps a lot.

Mhmm. It hasn't really made me, if you will, less busy. In a lot of ways, I become more busy because I'm able to do more work.

So I think that that's the second part to consider is the task versus the purpose of the job. This example really strikes home because my my sister-in-law, Erin, actually leads nuclear medicine at Stanford. Right?

So she's in radiology.

it doesn't result in layoffs. It results in us doing more more things. I met your new hire class today.

You seem to be hiring every week anyway. Yeah. That's exactly right.

Yeah. Right? The the more productive we are, the more ideas we can explore, the more growth as as a result, the more profitable we become, which allows us to pursue more ideas.

And so I think you're you're absolutely right that that if the job if if your if your life, if the world, the problems is literally already specified and there's no other problem to solve, then productivity would actually reduce the economy. But it's clearly gonna increase the economy. I think that the next part that I would consider is but, you know, people say, gosh.

All of these robots that we're talking about is gonna take away jobs. As as we know very clearly, we don't have enough factory workers. Our economy is actually limited by the number of factory workers we have.

Most people are are having a very hard time retaining their workers. We also know that the number of truck drivers in the world is severely short. And the reason for that is people don't want those jobs where you have to travel across the country and live in different parts of the world, different parts of the country, you know, every single night.

And so people wanna stay in their town, stay with their families. So I think the I think the first part is that having robotic systems is going to allow us to cover the labor shortage gap, which is really, really severe and getting worse because of aging population. This is this is not only United States, all over the world as you guys know.

And so we're gonna cover the labor shortage.

automation is gonna help us increase and solve the the the labor gap. Now people also don't don't remember that when we have cars, we need mechanics to take care of our cars. Mhmm.

And if you look at the robotaxis that are that are even on the streets today, it's taken ten years for that to happen. Look at all the maintenance crews and all of the the the various, you know, hubs that they're in where you have to take care of these robotaxis. And just imagine we have a billion robots.

Mhmm. It's going to be the largest repair industry on the planet. So I I think a lot of people don't they they just have to think through.

Mhmm. And this is the part where you said, when we create this type of automation, we create this other job. Right now, look at AI is creating so many jobs.

Mhmm. The AI industry is creating a boom of jobs.

I think a lot of policymakers have focused on, you know, we can't replace or reduce what we have when it's really there's far more demand in what we actually are not fulfilling. And in the case of lawyer, what's the what's the purpose of the lawyer versus the task of the lawyer? Mhmm.

Reading a contract, writing a contract is not the purpose of the lawyer. The purpose of the lawyer is to help you resolve conflict. And that's more than reading a contract.

It's more than writing a contract. The purpose is to protect you. That's more than reading a contract.

It's more than writing a contract.

interconnected, dependent networks of problems, this this, you know, big goop of a mesh of problems, it's always good to to go back and find a framework for what it is that we're talking about. In the case of AI, what is AI? Well, of course, the technology of AI and the capability of the capabilities of AI is about automation.

It's about automation of intelligence for the very first time. And you could combine it with mechatronics technology to embody that mechatronics and and make it perform tasks. Mhmm.

So that's what's AI automation. But what what is the stack that makes AI possible? What's the technology stack, the functional stack?

And, of course, the the the easiest way to think about that is is kind of like a five year five year five layer cake, which is at the lowest level is energy. Mhmm. It transforms energy to the output that I just described.

The next layer is chips. The next layer is infrastructure, and that infrastructure is both hardware, software. Right?

This is where land, power, and shell. This is where construction is. Data centers are.

The software stack Mhmm. You know, orchestrating the so it's software and hardware. The layer above that is where everybody thinks about, which is AI, which is the models.

We know this, but it's really helpful to understand that AI is a system of models, and AI is a technology that understands information. And there's human information. And so we oftentimes think about AI as a chatbot.

But remember, there's biological information. There's chemical information. There's physical information.

Information of all kinds. There's financial information. There's healthcare information.

There's information of all modalities, all kinds. AI is really, really broad. And, of course, human language is at the foundation of many things, but it's not the essence of everything because, as you know, biology molecules don't understand English.

They understand something else, right? Proteins don't understand English. They understand something else.

I think the next layer, the important thing is is that's where the AI models are, but there's a whole AI is very, very diverse. And then the the layer above that is is applications, and it depends on the industry. And you already mentioned open evidence that you mentioned Harvey.

There's Cursor. There's all kinds of right? There's all kinds of applications.

Full self driving is really an application, an AI application that is embodied into a mechanical car. Mhmm. And Figure is a AI application that has been embodied into a mechanical human.

And so so you got all these different applications. Well, this five layer stack is one way of thinking about it. And then the next way of thinking about, I just mentioned, is AI is really diverse.

When you now have this framework of what the the technology capabilities are, how to how to build the technology, and how diverse it is, then you can come back and think about, okay. Let's ask the question, how important is open source? Well, without open source, today, of course, the frontier models, the leading labs, have chosen to use a closed source application approach, which is just fine.

What people decide to do with their business models is really in the final analysis, their business. And they to calculate what is the best way for them to get the return on investment so that they could scale up and make better advances. However they made that calculus is fantastic.

On the other hand, without open source, as you know, startups would be challenged. Companies that are in in different industries, whether it's manufacturing or transportation or it could be in health care, without open source today, all of that AI work would be suffocating. And so they just need to have something that's pretrained.

They need to have some fundamental technology about reasoning. From that, they could all adapt, fine tune, you know, train their AI models into exactly the domain and application they want. And so what people really, really miss is just the incredible pervasiveness and the importance of open source to all of these industries.

Large companies Mhmm. Without without open source, some of some of 100 year old companies that I work with Mhmm. In in industrial spaces and health care spaces, they would be suffocated.

They wouldn't be able to do their work. Open source at this point is driving all of our data centers. It's driving a big chunk of telephony in the world in terms of Android or other devices.

It's driving a lot of the industrial applications. So it's already pervasive. Then I think the big question is Open source, without open source, higher ed.

Higher ed wouldn't happen. Education research. Mhmm.

Startups. I mean, the list goes on. Mhmm.

You know? And so so we talk we talk all day long about the tip, the most visible part of that, the most the part that's most newsworthy maybe. But underneath that is such an important space of open source AI.

And whatever we decide to do with policies, do not damage that innovation flywheel. So I spend a lot of time educating educating policymakers to help them understand whatever you decide, whatever you do, don't forget open source. Whatever you decide, whatever you do, don't forget biology.

we will have God AI.

When is that day? But but that someday that someday is probably on biblical scales, you know, I think galactic scales. I I think it's it's not helpful to go from where we are today to God AI.

Mhmm. And I don't think any company practically believes they're anywhere near God AI. And nor nor do I do I see any researchers having any reasonable ability to create God AI.

The ability to hunter understand human language and genome language and molecular language and protein language and amino acid language and physics language all supremely well. That got AI just doesn't exist. And and yet we have a lot of industries that need AI.

Mhmm. AI is, if if you will, at the simplistic level, it's just the next computer industry. Mhmm.

And give me an example of a company, an industry, a nation who doesn't need computers. Mhmm. And we all don't have to wait around for God AI for us to advance.

Right? So God AI is not showing up next week. I'm fairly certain of that.

Okay. That's right. AI God AI is not not gonna show up next year, but the whole world needs to move forward next week, next year, next decade.

I think that that the idea of a monolithic, gigantic company Mhmm. Country, nation state that has God AI is just It's unhelpful. To it's unhelpful.

It's too extreme. Then, in fact, if you wanna take it to that level, then we ought to just all stop everything. What's the point of having even governments?

I mean, why why why are they doing policies? God AI is gonna be smart enough to avert, you know, work around any policy. And so what's the point?

And so I I think that that we ought to bring things back to the ground, ground level, and start thinking about things practically and and use common sense.

with very well respected people who have who have painted a doom doomer narrative, end of the world narrative, science fiction narrative. And, you know, and I and I appreciate that that many of us grew up and and enjoyed science fiction. Mhmm.

But but it's not helpful. It's not helpful to people. It's not helpful to the industry.

It's not helpful to society. It's not helpful to the governments. Mhmm.

There are a lot of many people in the government who obviously aren't as familiar with as as comfortable with the technology. And when PhDs of this and CEOs of that Mhmm. Goes to governments and explain and describe these end of the world scenarios and extremely Mhmm.

Extremely dystopian future of the future, you have to ask yourself, you know, what is the purpose of that narrative, and what is their what are their intentions, and what do they hope?

You know, I I can't I can't guess what they what they have in mind. I know that the concern is regulatory capture. As a policy as a practice, I don't think companies ought to go to governments to advocate for the regulation on other companies and other industries.

Just in practice. Their their intentions are clearly deeply conflicted, And their intentions are clearly, you know, not completely in the best interest of society. I mean, they're obviously CEOs.

They're obviously companies. And obviously, they're advocating for themselves. And so I think if we can all come back to where are we today and think about where the technology is going to be, I mean, look, literally in one year's time, as we were talking about in the beginning, some of the most proud moments is when the industry was able to invest very aggressively in advancing AI technology instead of being slowed down.

Remember, just two years ago, people were talking about slowing the industry down. Mhmm. But as we advanced quickly, what did we solve?

We solved grounding. We solved reasoning. We solved research.

All of that technology was applied for good, improving the functionality of the AI, not you know Yet the end has not come. Yet the end has not come. It's become more useful.

It's become more functional. It's become able to do what we ask it to do. You know?

And so the first part of the safety of a product is that it performed as advertised. Mhmm. The first part of safety is performance.

That it's a it's supposed like, the first part of safety of a car isn't that some person is gonna jump into the car and use it as a missile. The first part of the car is it works as advertised. Mhmm.

99.999% of the time working as advertised. And so it takes a lot of technology to make that car or make that AI work as advertised.

And I'm really glad that in the last couple two, three years, the industry has invested so much in enhancing the functionality of the AI as advertised. And I think if if we were to to look at the next ten years, we have so much work to do to make it work as advertised. Meanwhile, as as, you know, you both of you invest so much in in the in the ecosystem, you see so many companies being built for synthetic data generation so that the AIs could be more grounded, more diverse, biased, more safe.

You're investing in a whole bunch of companies in cybersecurity using AI for cybersecurity. Right? People think that there's this AI.

The marginal cost of the AI is gonna go go down significantly, and it is. And, therefore, the AI is gonna be dangerous. It's exactly the opposite.

If the marginal cost of AI is going to go down significantly, that one AI is gonna be monitored by millions of AIs. Mhmm. And more and more AI is gonna be monitoring each other.

People don't can't forget that an AI is not going to be an agent by itself. It's likely the AI is gonna be surrounded by agents monitoring it. Mhmm.

estimated the cost of building the first ChatGPT, I think, Yeah. Versus now. I I you could do that on the PC now.

Yeah. Yeah. It's probably tens of thousands of dollars at this point or maybe even less.

Right? And so Yeah. It costs nothing.

Mhmm. And and He has an open source project that you can do in a weekend. Oh, is that right?

Okay. That's incredible. Right?

Yeah. We're talking about three years. Mhmm.

Mhmm. What people people said cost billions of dollars. Supercomputers built, raising billions of dollars in order to do all that now Mhmm.

Cost, you know, something that you can do on a weekend on a PC. And so that tells you something about how quickly we're making making AI more cost effective. Spark.

Sorry. Probably not quite a PC. Yeah.

Not quite a PC. Yeah. We're improving our architecture and performance every single year.

The first GBTU, I think, was trained on Voltus. Mhmm. And then Ampere.

You know? And and it wasn't I think the first breakthroughs, none of it included Hopper. Mhmm.

And, of course, Hopper last couple two, three years, and we're off in Blackwell for the last year and a half or so. And every single one of these generations, the architecture improves. And, of course, the number of transistors go up and the capacity goes up.

Every single generation, very easily, every every single year from a computing perspective, the combination of all that getting five to 10 x every single year it's not unusual. And here comes Rubin just around the corner. And so we're seeing five to 10 x every single year.

Well, compounded, it's incredible. Moore's law was two times every year and a half. And over the course of five years, it's 10 x.

Over the course of ten years, it's a 100 x. Mhmm. In the in the in the case of AI, over the course of ten years, it's probably a 100,000 to a million x.

Okay? And that's just the hardware. Mhmm.

Then the next layer is the algorithm layer and the model layer. The combination of all that, the fact that if you were to tell me that in the cost in the in the in the span of, you know, ten years, we're gonna reduce the cost of token generation by a billion times, I would not be surprised. Mhmm.

Okay? And so that's the tokenomics of of AI. On the training side, it's not quite as aggressive in in cost reduction, but it's close.

If you were to say that that every single year, we're increasing by two or three x over the course of ten years. Incredible.

and and this is where MOEs come in, as you know. The scale went up by a factor of 10, but the computational burden did not go up by a factor of 10. Because you're you're getting the compounded benefits of all three things.

The hardware is going up. The the algorithms of of training models are going up. And, of course, the model architecture is going up.

And we're getting the benefit of learning from each other. This is you know, let's face it. DeepSeek was probably the single most important paper that most Silicon Valley researchers read from in the last couple years.

It was the only thing that felt frontier that was open That's right. In years. That's right.

Because came out of the value of open source again. Yeah. Yeah.

Putting out these papers. Literally, DeepSeek benefited American startups and American AI labs Mhmm. All over.

And infrastructure companies. And infrastructure company all over. Probably the single greatest contribution to American AI last year.

Mhmm. And so if you said this out loud, of course, you know, people Mhmm. Kinda shutter that we're American AI is actually getting learning from and benefiting from AI from other nation.

But why would that be surprising? You know, AI researchers in all over America all over America are Chinese natives and come from different countries. We benefit from every country.

We benefit from every researcher. And, no, all of the world's ideas don't have to come from United States. And so I I think back to your your original question, it is the case that, you know, some of the narratives around around the cost of AI is about scaring everybody out of the market.

You know? Nobody ought to do pretraining but us. Nobody should do, you know, training these frontier models but us.

But because the because of innovation of models, algorithms, and the computing stack, the cost of AI is actually decreasing well more than 10 x every single year. And so if you're just one year behind or even six months behind, you could you could really stay close.

As the market grows, each one of these models could choose to have verticals Mhmm. Or segments where they wanna differentiate. Mhmm.

Somebody could decide to be a better coder. Somebody could decide to be just better at being easier to be accessible so that it could be a greater consumer product. Mhmm.

You know, the diversity of these models, as a result, you could you could probably make a niche leap without having to be great at everything else and still be super valuable to the market. Mhmm. It's no longer necessary to boil the entire ocean.

The fur two years ago, because it was called pretraining. Pre you know, people people said, well, you know, pretraining is over. First of all, pretraining is not over.

But the point of pretraining is to train yourself for training. That's why it's called pretraining, to prepare yourself to do the real training. And now we call it post training.

It's kinda weird. I I think it's just training. But pretraining is pretraining, and therefore, it's training.

Training, as you as as we all know, is is where compute scaling directly translates to intelligence. You you've you've largely now now this the the data the the data necessary to train the model is actually pretty small. Maybe it's just the verifiable results.

Now it's really algorithmic, very compute intensive. And so and you don't have to be good at everything in life, as you know, just like all of us. We don't we could decide because we don't have time to learn everything equally well.

We decide to choose a specialty and focus all of our energy on it. And we become superhuman or incredibly good at something that other people are not. And so I think AI ladders are gonna start doing the same.

They're gonna start bifurcating into various segments.

we are using we we use cursor here, and we use cursor pervasively here. Every engineer uses it. Mhmm.

And the number of engineers you just mentioned it. The number of people we're hiring today is just incredible. Yep.

Right? Monday is come to work at NVIDIA day. And why is that?

This is now the purpose and the task. Mhmm. The purpose of a software engineer is to solve known problems and to find new problems to solve.

Coding is one of the tasks. Mhmm. And so if the purpose is not coding, if your purpose literally is coding, somebody tells you what to do, you code it.

Alright? Maybe you're gonna get replaced by the AI. But most of our software engineers and all of our software engineers, their goal is to solve problems.

And it turns out we have so many problems in the company, and we have so many undiscovered problems. And so the more time they have to go explore undiscovered problems, the better off we are as a company. Nothing would give me more joy than if none of them are coding at all.

They're just solving problems. Mhmm. You see what I'm saying?

And so I I think that this framework of purpose versus task is really good for everybody to apply. For example, somebody who who's a waiter, their job is to not to take the order. That's not their job, as it turns out.

Their job is so that we have a great experience. And if somebody if some AI is taking the order, their job or even delivering the food, their job is still helping us have a great experience. They they would reshape their jobs accordingly.

And so so I think the the question about about cost of compute is really important. Let's let let me come back to one. The the reason why we are so dedicated to a programmable architecture versus a fixed architect remember a long time ago, a CNN chip came along, and they said NVIDIA is done.

And then and then the transformer chip came, and NVIDIA was done. People are still trying that. Yes.

Yep. And and the benefit of these dedicated ASICs, of course, it could perform a job really, really well. And transformers is a much more universal AI network.

But the transformer, as you know, the species of it is growing incredibly. The attention mechanism. The attention mechanism, how it thinks about context, diffusion versus autoregressive.

It's hybrid SSM transformers. Yeah. SSMs.

For example, Nematron. We just announced a new hybrid SSM. And and so the architecture of transformers is in fact changing very rapidly.

And over the next several years, it's likely to change tremendously. And so we we dedicate ourselves to an architecture that's flexible for this reason so that we can, on the one hand, adapt with remember, because Moore's law is largely over, transistor benefit is only tens 10% maybe a couple of years. Mhmm.

And yet we would like to have hundreds of x every year. And so the benefit is actually all in algorithms. And an architecture that enables any algorithm is likely gonna be the best one.

Right? Because the transistor didn't it didn't advance that much. And so I I think the the our dedication to programmability is number one for that reason.

We have so much optimism for innovation and algorithms and innovation software that we protect our programmability for that reason. The second thing is is by protecting this architecture, our installed base is really large. When a software engineer wants to optimize their algorithm, they wanna make sure that it doesn't run on just one this one little cloud or this one little stack.

They want it to run on as many mod on as many computers as possible. So the the fact that we protect our architecture compatibility, then flash retention runs everywhere. So SSMs run everywhere.

Diffusion runs everywhere. Auto regression runs everywhere. Just depending it doesn't matter what you wanna do.

CNN still run everywhere. LSTM still runs everywhere. And so the the this architecture that is architecturally compatible so that we have a large installed base, programmable for the future is really important in the way that we help to advance.

And as a result, all of this drives the cost down. And and I'm super proud that that our latest innovation, MBlink 72, we're the lowest cost token generation machine in the world by enormous amounts. And the reason for that is because MOEs are really, really hard.

I because of because of a couple two, three things. I think I think several industries are gonna are gonna experience their chatty PT moment. I believe that multimodality and very long context is going to enable, of course, really, really cool chatbots.

But the basic architecture, that in combination with breakthroughs in synthetic data generation is going to help create the ChatGPT moment for digital biology.

I see. Protein synthesis. I think we're good at protein understanding.

Now multi protein understanding is coming online, and we recently created a model called LaProteina. It's open. It's for multi protein understanding and represent representation learning and generation.

So so I think that the protein understanding is is advancing very quickly. Now protein generation is going to advance very quickly. ChatGPT moment proteins.

Yeah. There are a lot of interesting companies working on molecule design and end to end way like pie. Exactly.

And then and then, of course, chemical understanding and chemical generation. Mhmm. And then protein, chemical Yeah.

Confirmation, understanding, and generation. Mhmm. Is that right?

Synthetic data is gonna be really important because they just have such sparse, right, sparsity of data, and they just don't have as much as human language. And there, the the real breakthrough is going to be when we can train a a world foundation model, a foundation model for proteins, a foundation model for cells. I'm I'm very excited about both of those things.

Once we have a a foundation model, our understanding capability, our generative capability, that data flywheel is where we can take off. Mhmm. The the the the second area that I'm excited about, of course, reasoning made huge breakthroughs in language.

But because of reasoning, cars are going to be able to perform better. So instead of just perception cars and planning cars, they're gonna be reasoning cars. So these cars are gonna be thinking all the time.

And when they come up they come up to a circumstance they they've never encountered before, they can break it down into circumstances they have encountered before and construct a reason reasoning system for how to navigate through it. And so the out of domain, out of, you know, out of distribution Mhmm. Part of AI is going to very much be be addressed by reasoning systems.

The first era was smart sensors Mhmm. Connected into a car. Mhmm.

The Mobileye era. The Mobileye era. And and even even the very earliest days of of Weibo.

ADAS. Yeah. Yeah.

Even the earliest days of Weibo. The the you're taught you're using smart sensors, a lot of human engineered algorithms. Mhmm.

Available education. Severe mapping.

And so you so you're essentially creating a car that is driving on digital rails. Right? It's no different than than the rails at Disneyland.

It's like there are digital rails. And so that's the first generation. The second generation and during that generation, you have perception, world model, and planning.

Mhmm. And and these modules. And each one of these modules have the limits of their technology, and and perception was first input was was first affected by deep learning first, and then and then and then it propagated through the pipeline.

And so that but that system was too brittle. Mhmm. And it only knows how to perform what you taught it.

Mhmm. And now where we are are end to end models. And then and then where we're gonna go next are end to end models.

With racing. Yeah. There you go.

So that those are kind of the four eras. In a lot of ways, if we were to started self driving cars probably three years ago We'd still be those it would probably be exactly the same place. All our poor friends who were working in self driving.

Yeah. And and I don't I don't mind it. I've been working on on it for ten years.

NVIDIA's self driving car stack, by the way, number one rated safety in the world today.

We'll just jump straight much more optimistic with robotics because we we've kind of We've been through some foundational technology. You know, people are thinking about human robotics. Human robotics has a lot of challenges.

I mean, there's all the megatronics challenges. They're you know? Like, for example, it's not helpful if the robot weighs 300 pounds.

Mhmm. And what happens if it falls over and is interacting with kids and so on and so forth? And so so you got all kinds of challenges to deal with.

I'm certain that we're gonna we're gonna solve those. Mhmm. But remember, the fundamental technology that goes into a human robot robot can go into a pick and place robot.

Mhmm.

and surely a major one. But everything that moves will be robotic. Everything that moves will be robotic.

And everything that moves is a very large space. It's not all human or robot. And yet every AI will be multi embodiment, meaning, you know, just a human with our multi embodiment AI ourselves, we could sit in a car and embody that.

We could pick up a tennis racket, embody that. We could pick up a chopstick, embody that. And so we could embody the people.

General purpose. Right? That's true.

All these things. Exactly. And so AIs are gonna become general purpose.

You have one arm pick and place. Maybe it's two arms pick and place. It could be six arms pick and place.

You know? So so I think you're gonna have all kinds of different sizes and shapes. It could be a caterpillar.

It could be, you know, it could be an excavator. It could be all kinds of stuff. And so AI will embody those just as a just as a a construction worker embodies an excavator.

Mhmm. Embodies a tractor. You know, they you know?

Could there could there be a small number of companies then that do the embodiment for everything, or you're saying more there's gonna be niche applications? Definitely see a lot of software companies. And then those that software company could serve a lot of a lot of different verticals, but each one of the verticals will still have solution providers that then grounds it all, turns it into something that works perfectly.

Does it make sense? Yeah. Because in the case of AI for consumers, if it works 90% of the time, you're delighted.

You're you're, you know, you're mind blown. If it works 88% of the time, you're satisfied. In the case of most industrial and physical AIs, if it works 90% of the time, nobody cares about that.

They only care about the 10% that it fails. Mhmm. 100 basically, you know, 100% dissatisfaction.

And so you've gotta take it to 99.9999. So the core technology might be able to get get you to 99%.

Mhmm. And then a a vertical solution provider like a Caterpillar or somebody, they could take that core technology and make it 99.

for technology that is that is general purpose is that you don't have the r and d scale to build a general purpose technology. Now, of course, open source helps that tremendously Mhmm. Which is the reason why you're gonna see a, you know, a a big surge of vertical opportunities in AI in the next several years.

Mhmm. My my prediction would be over the course of the next five years, the excitement is going to be verticalization. Mhmm.

Notice we we're excited about OpenEvenus. We're excited about Harvey. We're excited about Cursor.

Cursor is is a horizontal, but it's kind of a horizontal vertical. Mhmm. You know?

And so I'm super excited about all the verticals. Mhmm. You know?

A lot of people said, yeah. AI is gonna get so God AI is gonna get so good that all these rapper companies are gonna be obsolete. It's just it misses the big point.

Mhmm. You know? The reason why you could talk about the reason why somebody can talk talk about somebody who's creating technology could talk about the life of a surgeon is because they've never been a surgeon.

The reason why somebody who builds an AI and talks about the life of an accountant and a tax expert is because they've never been a tax expert. You know? And so I think they just the reason why somebody could talk about being a busboy without being a busboy is because they never been a busboy.

And so I I think you you you've gotta be a little bit more empathetic about the depth of the complexity of the work and and try to truly understand the purpose of the work. Oftentimes, the the technology addresses the task. It doesn't address the purpose.

and we have enough energy to support AI? How do how do you think about that? On the first week of president Trump's administration, he said, drill, baby.

Drill. He there's so much flack for that. Mhmm.

If not for this entire change in in sentiment about energy growth in our country Mhmm. We can all concede now. We would have handed this industrial revolution to somebody else.

Mhmm. And we're still power constrained. We're still power constrained.

Yeah. Without energy, there can be no new industry. Mhmm.

And, of course, we've been energy starved now for, what, a decade? If not for the fact that president Trump reversed that narrative, we would be completely screwed. Mhmm.

Without energy, you can't have industrial growth. Without industrial growth, the the nation can't be more prosperous. Without being more prosperous, we can't take care of domestic issues.

We can't take care of social issues. You know? On and on and on.

And so the fact of the matter is we need energy to grow. We need every form of energy. We need, you know, natural gas.

We we need to be of course, we need more energy on the grid. We need more energy behind the meter. We're gonna need nuclear.

Wind is not gonna be enough. Solar is not gonna be enough. Let's just all acknowledge that we'll take it.

We'll take everything we can. But the fact that matters, I think, for the for the next decade Mhmm. Natural gas, you know, is probably the the only way to go forward.

for

It still drives climate innovation. Yeah. No question about it.

No question about it. And I I think that's exactly right that that, you know, doomer messages causes policy, and that policy may affect the industry in some way, but there's nothing more powerful than demand. Look at all the jobs that's being created.

Look at all the industries that's being formed around it. Sustainable energy likely.

it's too simplistic to say that everything that the doomers are saying are irrelevant. That's not true. A lot of very sensible things are being said.

It is too simplistic to say that when somebody is is optimistic Mhmm. That they're just naive. It means to be grounded in reality.

Yeah. Yeah. That optimistic people are just naive.

You know? And that that's obviously not true. But I think we just have to be mindful of the balance of it.

Mhmm. When 90% of the messaging is all around the end of the world and doom and the pessimism and, you know, I I think we we're scaring people Mhmm. From making the investments in AI that makes it safer, more functional, more productive, and more useful to society.

And so we just, you know, more secure. You know, all of that takes technology. Security takes technology.

Safety takes technology. I appreciate that my car is safer today because it has better technology than a car fifty years ago. Mhmm.

And so so I I think it takes technology to be safe, technology to be secure. And so I I'm I'm I'm I'm delighted to see that the the advancement of technology is still accelerating and ongoing. And so we just have to make sure that the the policymakers around the world, the governments are able to are are thinking about balancing these two ideas.

that that our relationship with China will improve. Mhmm. That president Trump and the administration has a really, really grounded and common sense attitude about and philosophy around around how to think about China.

That that they're an adversary Mhmm. But they're also also a partner in many ways. And that the idea of decoupling is naive.

And the idea of decoupling for whatever reason, philosophical reasons or national security reasons, it's just not not it's not based on any common sense. And the more you the more deeply you look into it, the more the two countries are actually highly coupled. Mhmm.

Both countries ought to ought to invest in their own independence. You know, when you depend too much on someone, the relationship becomes too emotional, as you know. And so it's good to have some independence and or as much independence as either either would like, but to recognize that there's a lot of coupling, a lot of deep dependence between the two countries.

And and I think there's a there needs to be a nuanced strategy, a nuanced attitude about how to how to how to manage this relationship in a productive way for all of the people of two countries and for all of the people around the world. Everybody depends on a productive, constructive relationship of the two most important nations and the single most important relationship for the next century. And so we have to find that answer, and and I'm I'm just really delighted that president Trump is looking for a constructive answer.

And so I I think that next year will be a much better better better year than the last several. I'm happy with the administration was able to to to suggest a a an export control policy that is grounded on national security, recognizing that they already make so many chips themselves, and they they can depend on Huawei themselves for their military, for their national security. They got ample technology to do that.

And so that American technology, although general purpose, is unlikely to be used by their military because their military is too smart, just as our military is too smart to to use their technology. And so it's grounded on national security. It's grounded on on technology leadership.

It's grounded on national prosperity. You know, one of the things that that we just always have to remember is that the world's mightiest military is supported by the world's mightiest economy. And so the wealth that we generate brings jobs home, creates prosperity in The United States, provides for tax revenues, and ultimately funds the mightiest military on the planet.

And so that circular system, that interconnected system requires a nuanced strategy. Mhmm. And and and and I'm I'm I'm pleased to to see some of the progress in that area that allows American technology companies to keep America first and keep America ahead and to to support American technology leadership, on the one hand to win globally.

Mhmm. And and then and then China, of course, is sorting itself out. You know?

I mean, not not sorting, but they're sorting out the attitude about how to think about American technology.

between the two countries. The way that I would think through that is go back to the first principles of technologies again. Mhmm.

And let's say the Internet. You have the chip industry. You have the systems in industry, the software industry.

You have the services industry on top. Remember, China's Internet growth has been a boom for Intel and AMD selling CPUs. Mhmm.

Micron selling DRAMs. SK Hynix and Samsung selling DRAMs. Mhmm.

It is the second largest Internet market for American technology industry. Mhmm. And so so maybe Mhmm.

Maybe it wasn't helpful to some layer of the stack. Mhmm. The Googles of the world.

That's right. Uber. But don't don't exclude every layer of the stack.

Always come back. Every single one of these things, take a step back, and look at the whole stack. Maybe that's the theme for today as well, and it makes sense that you would you would send this message.

the whole stack. And remember, as as as Intel and AMD prospered with the Internet industry in China grow the China industry growth, Don't forget, China also contributed tremendously to open source. No country in the world contributes more to open source than China.

And look at all the startups here in America that were able to benefit from all of that open source to create the new startups that are here. And so you can't look at one area in isolation. You have to look at the whole life cycle of the technology and look at every layer of the stack.

Doesn't make sense? When you take a look at that from that lens Mhmm. China's Internet industry generated enormous prosperity for America.

Just not at the Internet company per se.

on the business side. Are we in an AI bubble? AI bubble.

Yeah. There's a lot of ways to reason through that. And so so, again, you know, when when asked that question, my mind goes to what is AI, and where are we in that?

There's AI. Then there's computing. You know, as you know, NVIDIA invented accelerated computing.

Accelerated computing does computer graphics and rendering. AI doesn't. Accelerated computing does data processing.

SQL data processing. AI doesn't. Mhmm.

Accelerated computing does molecular dynamics and quantum chemistry. AI doesn't. You know, all these are all things that people could say someday AI will, but it doesn't today.

Accelerated computing is really essential for classical machine learning, XGBoost, recommender systems, the whole process of feature engineering, extract, load, and transform. That entire data science machine learning's life cycle. Accelerated computing is used for all of that.

The first thing to go to is in the context of NVIDIA. What we see is the the the dynamic is a shift from general purpose computing to accelerated computing because Moore's Law is largely ended. You can't use CPUs for everything anymore like you used to, and so it's just no longer productive enough.

It's not deflationary enough. Mhmm. And so so we have to move towards a new computing model, and that's where Accelerated comes in.

If you if generative AI well, excuse me. If chatbots let's just go, you know, OpenAI and Anthropic and Gemini. If none of that existed today, NVIDIA would be a multi $100,000,000,000 company.

And the reason for that is because, as you know, the foundation of computing is shifting to accelerated computing. That's the first thing to to realize is is to take a step back and ask yourself what is actually happening. Now the next layer up.

The question about AI now becomes what is AI? Now we ask that we ask the AI bubble question, and we always go back to OpenAI's revenues. A 100%, don't we?

Mhmm. You ask somebody, hey. Is there an AI bubble?

Everybody goes directly to OpenAI's revenues. First of all, if OpenAI currently has twice the capacity, their revenues would double. You guys know that.

If they have 10 times the capacity, they're I really believe their revenues were 10 times. And so they need capacity. This is no different than NVIDIA needs wafers from TSMC.

Just because, you know, NVIDIA exists and and we're doing great, doesn't mean we don't need capacity. We need capacity. We need capacity of DRAM.

So in our world, it's sensible to everybody. We need capacity. Well, in their world, they need factories.

And if they don't have factory capacity, how do they generate tokens, which is where we started our conversation today. And so they need factory capacity in order to increase their revenue growth. But nonetheless, we also said that AI is more than chatbots.

It includes all these different fields of science. NVIDIA's AV business is coming up on $10,000,000,000. Nobody ever talks about that.

And you have to train world models. You have to train these AVs, and it's happening robotaxis happening all over the world. Our AI work with digital biology.

Our AI work in financial services. The whole industry of quants, quantitative trading is moving towards. Yeah.

There used to be classical machine learning, a whole bunch of human featured they call quants. Right? These these specialized mathematicians were trying to figure out what the predictive features are.

Now we use AI to figure it out. And so in order to have instead of having quants, you need a lot of supercomputers. Financial services is one of our fastest growing segments.

Billions of dollars in in quants, you know, in financial services. Billions of dollars in AV. Billions of dollars in robotics coming up.

Billions of dollars in digital biology. And so how big can that all that be? Well, simple logic as this.

Simple math. Whether you you think that AI is going to replace shortage, labor shortage, or workforce shortage in any kind, let's ignore that for a second. The world is at a $100,000,000,000,000 in GDP.

Out of that, let's just say 2%, 2% annually is R and D. And let's just go back in time. Five years ago, if you were to take the largest drug discovery company in the world, drug company in the world, and where's all of their R and D?

Wet labs. Today, what are they doing? Building supercomputers.

And so there's a fundamental shift in how they think about that $2,000,000,000,000. It used to be $2,000,000,000,000 for the old way of doing things. It's now gonna be $2,000,000,000,000 in the AI way of doing things.

Well, $2,000,000,000,000 is gonna need $2,000,000,000,000 of R and D is gonna be powered by a whole bunch of infrastructure. And that's the reason why we're building supercomputers everywhere around the world. And so so I think if you if you reason about it from the outside in, you know, either from the foundation up, from the outside in, you come to the conclusion that what we're experiencing, what all three of us are experiencing, which is the amount of computing demand is insane.

Mhmm. Give me an example of a startup company that goes, no, we're good. They are all dying for computing capacity.

Give me an example for a researcher in any university, a scientist in any company who says, got plenty of capacity. Everybody is dying for capacity. And so we have a global, multi company, multi industry shortage.

It's not just about OpenAI, even though OpenAI could use a lot more capacity as well. So I think I think how we think about this with with the narrative, the narrative is not helpful, and it's a little bit too superficial to say, how do you prove there's an AI bubble? $12,000,000,000 of revenues, hundreds of billions of infrastructure being built.

It's a little bit too simplistic.

things that get a lot of attention, but then you map it against what's actually happening. Yeah. And the growth of these companies using AI, and it's just a completely different world.

And and and if you wanna find out where the world's innovation is happening Mhmm. I would not go find out at an enterprise. Mhmm.

Would you guys agree? Yeah. Enterprise is like the slowest adopters of new technologies.

I would go talk to all of the startups, the thirty, forty thousand startups that are currently doing this stuff. I would go talk to OpenEvidence. How how's it working?

I would go go talk to Cursor. How's coding working, by the way? You know, I would just go talk to these people.

conservative industries. Right? Like health care or, you know, skeptical industries like engineering.

Health care. The most, Right? The most conservative of all.

But guess what? They are so concerned about getting the right answer Mhmm. That the ability to have something like open evidence Yeah.

To do grounded research, high quality research, and get that get that research as information to you. Nobody wants to do research. They want answers.

Nobody wants to do search. They want answers. Is that right?

Abridges is a great example of that too, they're basically making it really easy to do the physician notes instead of the physician sitting there and doing it.

Then Just like archive papers. Mhmm. Yeah.

There was a time you you so you and I both tried to read and read and Yeah. Both used to do. Mean, I don't do that anymore.

But here I still try. Now now I just load it all into ChatGPT. Mhmm.

Mhmm. You know? Now I just load it all in with all of the the ones that are interesting, and and then I make it learn it.

Yeah. And then I, you know, make it summarize. And you plug that another summary, and I I interact with it.

But but the point is we used to do search. We don't do it anymore. I don't do search.

We used to do research. Mhmm. You know?

The goal is to get answers. The goal is to get smarter, and these AIs allow us to help us do all that. And I think all of it all of it comes back it's all more helpful if you come back to the framework that says AI is a multilayer cake and that AI is not just a chatbot.

AI is very, very diverse in all of the industries and modalities and information and applications that it addresses. When you think about wanting to win, that America should win AI, it should not just be America should have this company win AI, but we should try to win across the board. And across domains.

Across domains. Exactly. And when we think about open source, all of a sudden, this is a helpful framework.

When we think about winning, it's a helpful framework. When we think about energy is a helpful framework that because we need factories, factories need energy. And without energy, we have no factory.

Without factories, we have no AI. That's a helpful framework. And so I think if if we if we have a better understanding, a system, a framework for understanding what AI is, I think the narratives will become more common sense.

The narratives will become more pragmatic Uh-huh. Become more balanced. We wanna keep people safe.

But one of the best ways to keep people safe is advancing a technology quickly. Mhmm. Mhmm.

And and I think the industry is doing that, and I'm very proud of the industry for doing that. No one wants to drive a car from, you know, the first decade of cars. And so I think ABS is a really good thing.

Yes. ABS is a really good thing. Lane keeping is a really good thing.

There's no question FSD is a really good thing. And I think people will be excited about the, you know, third or fourth year of AI. Yeah.

No no doubt. And and I I say with great pride that the industry made tremendous strides this last year. Mhmm.

All the technologies we've mentioned and that the scaling laws are so intact that we we now know that more compute, more intelligence. Mhmm. And and, gosh, the the the innovations in one in in one sector diffuses and spreads across all of the other sectors so fast.

I'm so happy to see all that. And so I think the next five years, it's gonna be extraordinary. No no doubt about it.

And I think next year is gonna be incredible. Amazing. Well, we're excited to talk to you at the end of next year too.

Yeah. Looking forward to it. Thanks so much, for all the work that you guys do.

Congratulations. What a great year. Wow.

It's for some amazing year. Yeah. Elad, thank you.

Yeah. Thank you. Happy New Year.

Happy New Yeah.