Jensen Huang – Will Nvidia’s moat persist?

Jensen Huang · 1h 43m · youtube.com ↗

24 korrents from this recording

Jensen Huang did not write this page.

Every claim below is a statement made in this recording, quoted word for word and linked to the second it was said, so you can hear it rather than take our word for it. The wording comes from the transcript published alongside the recording; the sentence above each quote is our reading of the claim, not their wording.

0:01:47 · watch on youtube.com↗

Our reading
Nvidia's business is the transformation of electrons into tokens, and its job is to do as much as necessary and as little as possible in between.

Their wordsmy mental model of our company the input is electron the output is tokens that is in the middle Nvidia and our job is to to do as much as necessary as little as possible to enable that transformation to be done at incredible capabilities

Interesting
2 min later
0:03:18 · watch on youtube.com↗

Our reading
AI will not commoditise software tools; it will multiply them, because every new agent is another tool user.

Their wordsI actually see the opposite of what people see I think the number of agents are going to grow exponentially. The number of tool users are going to grow exponentially and it's very likely that the number of instances of all these tools are going to skyrocket.

Interesting
2 min later
0:05:47 · watch on youtube.com↗

Our reading
Nvidia's suppliers make their investments for Nvidia and nobody else because only Nvidia's downstream demand can absorb what they build.

Their wordsNow why are they willing to make the investments for me and not someone else and the reason for that is because they know that I have the capacity to buy it buy their supply and sell it through my downstream.

Interesting
8 min later
0:13:44 · watch on youtube.com↗

Our reading
The doomers said radiology would be the first career to go; a decade later the world is short of radiologists.

Their wordsYou know, radiology is is going to be the first career to go. Nobody's the world's not going to need any more radiologists. Guess what? But we're short of radiologists.

Interesting
1 min later
0:15:00 · watch on youtube.com↗

Our reading
No bottleneck in the AI supply chain lasts longer than two or three years.

Their wordsAnd so none of that my point is that none of the bottlenecks last longer than a couple 2 three years. None of them.

Interesting
3 min later
0:17:37 · watch on youtube.com↗

Our reading
Nvidia did not build a tensor processing unit — it built accelerated computing, whose market reach no ASIC can match.

Their wordsOur market reach is far greater than any any TPU can any ASA can possibly have. And so if you look at our position, uh we're the only company that that accelerates applications of all kinds.

Interesting
0:17:59 · watch on youtube.com↗

Our reading
A homebuilt accelerator forces you to be your own operator; Nvidia systems are designed for anyone to operate, which is why they are in every cloud.

Their wordsbecause our computers are designed to be operated by other people, anyone who's an operator could buy our systems. Most of these homebuilt systems you have to be your own operator because it was never designed to be flexible enough for other people to operate.

Interesting
4 min later
0:22:00 · watch on youtube.com↗

Our reading
Moore's law now yields about 25% a year, so 10x and 100x leaps can only come from changing the algorithm and how it is computed, every single year.

Their wordsTPUs like anything else is impacted by Moore's law. And we know that Moore's law is increasing about 25% per year. And so the only way to really get 10x leaps, 100x leaps, is to fundamentally change the algorithm and how it's computed every single year.

Interesting
10 min later
0:31:35 · watch on youtube.com↗

Our reading
Nvidia's engineers can routinely get another 2x out of an AI lab's stack, and doubling the tokens a fleet produces doubles that lab's revenue.

Their wordsAnd so our expertise um helps our our our um uh our AI labs partners get another 2x out of their stack easily. Often times it's not unusual that we you know by the time that we're done optimizing their stack or optimizing a particular kernel their model sped up by 3x 2x 50%.

Interesting
1 min later
0:32:17 · watch on youtube.com↗

Our reading
No platform in the world has a better performance-per-total-cost-of-ownership ratio than Nvidia's, and nobody will run the public benchmark to show otherwise.

Their wordsNvidia's computing stack is the best performance per TCO in the world, bar none. Nobody can demonstrate to me that any single platform in the world today has better performance TCO ratio. Not one company.

Interesting
5 min later
0:37:03 · watch on youtube.com↗

Our reading
Anthropic is a unique instance and not a trend: without it there would be essentially no TPU or Trainium growth at all.

Their wordsanthropic is is an is a unique instance um and not a trend. Uh without an anthropic, why would there be any TPU growth at all? It's 100% anthropic. Without anthropic, why would there be any tranium growth at all? It's 100% anthropic.

Interesting
1 min later
0:38:07 · watch on youtube.com↗

Our reading
Building an ASIC is not the hard part; building something better than Nvidia is, and that is not easy.

Their wordsJust because you're going to build an ASIC, you still have to build something better. than Nvidia. And it's not that easy building something better than Nvidia.

Interesting
1 min later
0:38:56 · watch on youtube.com↗

Our reading
Moving to a custom ASIC does not escape the margin — the ASIC vendor’s margins are nearly as high as Nvidia’s.

Their wordsAnd so so I think the the ASIC margins are are incredibly good from what I can tell and and they believe it. They believe it so too. And so they're they're quite proud of their their incredible ASIC margins.

Interesting
1 min later
0:39:59 · watch on youtube.com↗

Our reading
Nvidia's real miss was not seeing that a frontier lab could never be funded by venture capital, so its compute suppliers had to become its investors.

Their wordsNor nor did I I would say my mistake is I didn't deeply internalize that they they really had no other options that that that a VC would never put in 510 billion of investment into an AI lab with the with the hopes of it turning out to be anthropic. And so that was my miss.

Interesting
4 min later
0:44:23 · watch on youtube.com↗

Our reading
Nvidia should build only what would not otherwise exist — twenty years of CUDA at a loss is the test, and the world already has plenty of clouds.

Their wordsthe work that we do with building our our computing platform. If we don't if we don't do it, I genuinely believe it doesn't get done. If we didn't take the risk that we take, if we didn't build MVLink the way we built, if we didn't build the whole stack, if we didn't create the ecosystem the way we did it, if we didn't dedicate ourselves to 20 years of CUDA while losing money most of that time, if we didn't do it, nobody else would have done it.

Interesting
3 min later
0:47:48 · watch on youtube.com↗

Our reading
Nobody can pick winners: Nvidia's own first graphics architecture was precisely wrong, and it would have topped the list of companies expected to fail.

Their wordsNvidia would be the top of that list not to make it. You know, this is long before you, but Nvidia's graphics architecture was precisely wrong. It's not a little bit wrong. We created an architecture that was precisely wrong.

Interesting
6 min later
0:54:17 · watch on youtube.com↗

Our reading
Allocating scarce chips to the highest bidder is bad business — you set your price and keep it, however high demand goes.

Their wordsBecause it's it's a bad business practice. You you set your price. You set your price and then and then people decide to buy it or not. And and um uh there there I I understand that that others in the chip industry um uh change their prices when demand is higher. Uh but we just don't we just don't that's just never been a practice of ours.

Interesting
2 min later
0:55:58 · watch on youtube.com↗

Our reading
You can bet your entire business on Nvidia shipping a better architecture every single year and your token cost falling by an order of magnitude — no ASIC team can promise that.

Their wordsPick your ASIC team where you can say I can bet the farm of I can bet my entire business that you will be here for me every single year. Your cost, your token cost will decrease by an order of magnitude every single year. I can count on it like I can count on the clock.

Interesting
3 min later
0:59:05 · watch on youtube.com↗

Our reading
Export controls cannot deny China compute: it already manufactures most of the world’s mainstream chips and has some of the world’s best computer scientists.

Their wordsAnd so you just have to first realize that chips exist in China. They manufacture 60% of the world's mainstream chips, maybe more. It's a very large industry for them. They have some of the world's greatest computer scientists.

Interesting
11 min later
1:10:09 · watch on youtube.com↗

Our reading
Most advances in AI have come from algorithms rather than raw hardware, which makes China's army of AI researchers its fundamental advantage.

Their wordsWe have got to acknowledge that most of the advanc advances in AI came out of algorithm advances not just the raw hardware. Now if most advances came from algorithms and computer science and programming tell me that their army of AI researchers is not their fundamental advantage.

Interesting
11 min later
1:20:46 · watch on youtube.com↗

Our reading
Computing is not a car market: x86 and ARM persist because ecosystems cost too much time and energy for anyone to replace.

Their wordsComputing is not like that. There's a reason why the x86 still exists. There's a reason why ARM is so sticky. These ecosystems, these ecosystems are hard to replace. It costs an enormous amount of time and energy and most people don't want to do it.

Interesting
12 min later
1:32:32 · watch on youtube.com↗

Our reading
Blackwell is 50 times Hopper while the transistors themselves improved 75% in three years — the gain is architecture, not lithography.

Their wordsBetween Hopper and Blackwell from the transistors themselves, call it 75%. It was 3 years apart. 75%. Blackwell is 50 times hopper. My point is architecture matters.

Interesting
5 min later
1:37:10 · watch on youtube.com↗

Our reading
Nvidia does not run rival chip architectures in parallel because it simulates them and they come out provably worse.

Their wordsYeah, we we could do all of those things. Um it's just not better. And we simulate it all. they're in our simulator provably worse and so we wouldn't do it.

Interesting
4 min later
1:40:45 · watch on youtube.com↗

Our reading
General-purpose computing has largely run its course; domain-specific acceleration is the way computing keeps scaling.

Their wordsthe ability for general purpose computing to continue to scale has largely run its course and the only the the not the only way but the the way to do that is through domain specific acceleration

Interesting