Jensen Huang: The Doomer Hoax, Superintelligence is Here, and The Future of AI (ft. President Trump) — Transcript
Full transcript
- 0:00Some people call it vision. Vision's an
- 0:03awfully big word to me because I I
- 0:04believe, first of all, vision matters.
- 0:07>> [music]
- 0:09>> We preempted the weekly show, and
- 0:12there's only three people we preempt the
- 0:13show for.
- 0:15President Trump, Jesus, and Jensen.
- 0:17[laughter]
- 0:18>> The number one podcast [music]
- 0:19in the world.
- 0:20>> That's Jensen Huang. He's the founder,
- 0:22president, and CEO of Nvidia.
- 0:23>> Whether you know it or not, his
- 0:24decisions are shaping your future.
- 0:26>> Nvidia is the most important stock in
- 0:28this market, and Jensen is arguably the
- 0:29most
- 0:29>> best executive in history.
- 0:31>> Revenue [music] exploded 97%
- 0:34year-over-year.
- 0:34>> Not only is demand already strong, it's
- 0:36actually accelerating.
- 0:38>> Nvidia is the only computing platform
- 0:40that is a full-stack AI factory. [music]
- 0:42A GPU is like a time machine
- 0:44because it lets you see the future
- 0:46sooner. And if we could see the future,
- 0:48and we can predict the future, then we
- 0:50have a [music] better chance of making
- 0:52that future the best version of it.
- 0:55>> Please welcome Jensen Huang.
- 0:58>> [music]
- 1:01>> Oh, we got a stand-to-go on the way in.
- 1:04>> Oh, come on.
- 1:05>> Stand-to-go.
- 1:07Stand-to-go on the way in.
- 1:10That's our guy.
- 1:12>> [laughter]
- 1:12>> Ladies and gentlemen, GPU Jesus.
- 1:19They love you.
- 1:20>> They love you.
- 1:20>> Thank you. I love you back.
- 1:22>> We like the new jacket.
- 1:22>> Number one podcast in the world.
- 1:26>> Wow.
- 1:27>> We like the new jacket.
- 1:28>> Well,
- 1:29>> You auctioned the old
- 1:30>> I just I felt you guys needed some
- 1:32energy.
- 1:33>> Yes. This is the right time.
- 1:34>> talking about serious stuff here, but we
- 1:36need to talk about it with energy.
- 1:38>> Yes. Let's let's start with this
- 1:41essay from this weekend.
- 1:43>> Which one?
- 1:44>> [laughter]
- 1:46>> Let's start with Dario's essay because I
- 1:47>> Was Hemingway involved?
- 1:49>> [laughter]
- 1:50>> No.
- 1:50>> Actually, did anybody run it through
- 1:51Pangram? I don't even know how much of
- 1:53it was AI helped, but that was a pretty
- 1:57incredible thing, and then, I think what
- 1:59a lot of people were surprised by was
- 2:01the coalescing of the Frontier Labs
- 2:03around the SA itself. Just, Jensen,
- 2:05unpack what happened, how you read it,
- 2:08how you interpreted it, and then we'll
- 2:10get into some details that were inside
- 2:11of it. But, maybe just the high-level
- 2:13thoughts to kick it off.
- 2:14>> Well, first of all, there were a lot of
- 2:15stuff in there.
- 2:16>> Yeah.
- 2:17>> And and
- 2:19uh the first there is a there's a part
- 2:20about about safety, which we have to
- 2:23take very seriously. Safety's paramount,
- 2:25obviously. Um uh safety and leadership
- 2:28are not false they're false choices.
- 2:31You're you're able to innovate quickly.
- 2:34You're able to execute quickly. And and
- 2:36America's able to lead and to do it
- 2:39safely. I think those are those are
- 2:40false choices, but safety's obviously
- 2:42important. Uh there's a matter of of
- 2:46internal control that I think he was
- 2:48speaking to. Uh obviously, the the
- 2:50Coxswain uh whistleblower is very
- 2:52serious matter. When whenever you have a
- 2:55whistleblower, you have to take it very
- 2:56seriously. I thought
- 2:58Coxswain [clears throat] had great
- 2:58courage uh to uh put out put out
- 3:02uh what his concerns were. Um and even
- 3:04then, there were some issues that were
- 3:06kind of conflated within that. Uh I
- 3:08think the the the whistleblowing is is
- 3:10fine. I think the the uh scientific
- 3:13prediction
- 3:15uh uh about the future uh is less fine
- 3:18because it's not grounded on science,
- 3:20obviously. And um uh it it was expressed
- 3:23by a scientist, but it was obviously not
- 3:24grounded on science. And and so, I take
- 3:26I take issue with that. But, obviously,
- 3:28the whistleblower part of it, uh
- 3:30>> [snorts]
- 3:30>> you know, I I think there's just a whole
- 3:31bunch of stuff. Uh
- 3:33pausing uh
- 3:35uh pacing uh
- 3:38uh those are all the voluntary things
- 3:39that they could do if they feel that
- 3:41their company is out of control.
- 3:43Uh if Coxswain saw something, uh you
- 3:45know, obviously, we didn't we don't know
- 3:47what Coxswain saw. But, he if he saw
- 3:49that the company was out of control
- 3:52and maybe it's a transition from uh
- 3:56research to engineering. As you know,
- 3:58these labs are research are
- 4:00transitioning from research to
- 4:01engineering. Extraordinary talent,
- 4:03extraordinary engineering. Um but
- 4:05obviously engineering is different than
- 4:06research. Maybe that transition is
- 4:08clumsy. You know, I don't We don't know
- 4:10what what he saw and ultimately only he
- 4:12knows. Um but if there was a matter of
- 4:15lack of control, that's a different
- 4:16topic. Um how should the government deal
- 4:18with it? Now all of a sudden uh
- 4:20regulation and I mean it just covers
- 4:23everything in one block.
- 4:24>> just help us sort of unpack? We we play
- 4:26We tried to play this game actually this
- 4:27week on the pod and it was difficult,
- 4:29which is
- 4:30how do you describe like, you know, my
- 4:32mom calls me and she's like, "Chamath,
- 4:33what is this whole
- 4:35civilizational death thing?" I don't
- 4:37know how to explain it to her. So, when
- 4:39you have very smart people like that,
- 4:41quantize it and quantify it. I think
- 4:43that's probably what's perturbing to
- 4:45some people. They're like, "What does
- 4:46that mean? 10% of extinction?"
- 4:48Nobody knows how to explain that to the
- 4:50average person, how that's even
- 4:52possible.
- 4:53>> Well, first of all, we shouldn't.
- 4:55Uh because it's made up. Uh first of
- 4:57all, I THINK THAT
- 4:59>> [laughter]
- 5:01[applause]
- 5:03>> WE SHOULDN'T BECAUSE IT'S MADE UP. AND
- 5:05THESE ARE These are well-educated.
- 5:08Uh they're called researchers. Um
- 5:11obviously they're working a lab. And so,
- 5:14the confluence of these words and then
- 5:17and then and then the prediction is
- 5:18alarming and troubling and it shouldn't
- 5:20be done. It's It's It's irresponsible.
- 5:23Now, the fact of the matter is, let's go
- 5:24back and look at the real facts. The
- 5:25facts are uh there was a prediction that
- 5:28in 5 years' time, radiology will be
- 5:31completely taken over by artificial
- 5:32intelligence and there'll be no
- 5:33radiologists in the world. That has
- 5:35proven to be exactly the opposite. We
- 5:37need more radiologists than ever in the
- 5:39world. However, AI has taken over
- 5:41radiology completely, which is great. It
- 5:43has automated scan reading, which is
- 5:45great. Um uh, there was a prediction
- 5:48that within 6 to 12 months, wasn't it
- 5:50just last year, within 6 to 12 months,
- 5:53uh, 90% of code would already be
- 5:56generated, uh, by AI. That has turned
- 5:59out to be wrong. Uh, within 6 to 9
- 6:01months, that was predicted last year,
- 6:0350% of entry jobs will be wiped out.
- 6:05That has proven to be wrong.
- 6:07Uh, let's see, what else what else has
- 6:08proven to be wrong? I mean, all of these
- 6:10predictions have been wrong.
- 6:12>> Right.
- 6:13>> And if you're a
- 6:14>> that GPT-2 would be too unsafe to
- 6:16release, that Llama 3 would be too
- 6:18unsafe to release.
- 6:20>> Yeah, we've heard these.
- 6:21>> Half of white-collar jobs would be gone
- 6:23next year.
- 6:23>> The jobs apocalypse, yeah.
- 6:25>> Yeah.
- 6:25>> We have to take accountability. We have
- 6:28to take account for all of the stupid
- 6:30predictions that were made.
- 6:32>> Right.
- 6:33>> Somebody has Somebody has to take
- 6:36>> No.
- 6:36>> [applause]
- 6:37>> And
- 6:38And so, we ought to just keep track of
- 6:39all that, and of course, people do, and
- 6:42remind us that those those predictions
- 6:45are inconsistent
- 6:47with
- 6:48ultimately, America winning the AI race.
- 6:50>> The short form for that is some people
- 6:52are saying, you know, they say, "Trust
- 6:54the experts." [clears throat] And they
- 6:55use the analog of COVID, which again,
- 6:58started with people that were
- 6:59researchers, educated people that had an
- 7:01asymmetric awareness of the thing that
- 7:03the rest of us did not, saying things
- 7:05that ultimately turned out, we find out
- 7:07in facts, uh, not to be true. Um, and
- 7:10so, there's this war that's happening
- 7:12right now between the trust the experts
- 7:15movement and the, you know, "Well, let's
- 7:17just look at the actual history of these
- 7:18predictions, and let's just think more
- 7:20methodically."
- 7:22Where is this coming from? Because it's
- 7:24coming from inside the places that's
- 7:26actually making it. Like, what do you
- 7:27think is the psychological makeup, or
- 7:29what is the real incentive? Maybe it's a
- 7:31business incentive. Maybe it's a
- 7:32political incentive. Can you just maybe
- 7:34guess, or how do you how do you think
- 7:36about what why they're doing this?
- 7:38>> Well, first of all, I got to tell you,
- 7:39these [clears throat] are some of the
- 7:40most consequential companies in history.
- 7:43Uh
- 7:43uh, extraordinary engineers,
- 7:45extraordinary researchers, really
- 7:47fantastic work. Um
- 7:49>> [clears throat]
- 7:50>> Uh on the one hand, uh I work very
- 7:52closely with them as companies to
- 7:53companies.
- 7:54Uh
- 7:55on the other hand, uh we have to have
- 7:57conversations like this in public. And
- 8:00it's really unfortunate. And I I think
- 8:02that that these these companies um
- 8:05really ought to be built the way that we
- 8:07used to build companies, which is in
- 8:09silence.
- 8:10>> Right.
- 8:11>> You know, and
- 8:13>> So wait, Jensen, you don't allow anybody
- 8:16in your organization to speak for the
- 8:17entire organization, especially when
- 8:19they're having like a bad weekend or
- 8:21they rage quit? They're they're not
- 8:22allowed to tweet on your behalf and the
- 8:24organization's behalf?
- 8:25>> Uh no, because well, that's that's what
- 8:28they decided when they came to work for
- 8:29us. And we told them, uh these are this
- 8:32is the way you behave when you work in
- 8:33our company. And and if you would like
- 8:36if you like the culture of our company,
- 8:38um uh which as you know, the Nvidia
- 8:41culture and the Nvidia employee base,
- 8:43incredibly happy.
- 8:44>> Yeah.
- 8:44>> Uh they like the fact that the company
- 8:46is consistent, that we're stable, that
- 8:48our core values are consistent with
- 8:51taking care of the families and creating
- 8:52the conditions by which they can do
- 8:54their life's work. Um
- 8:56that we do meaningful work. We do it We
- 8:58do it as quietly as we can. And we
- 9:01contribute to everybody else's success,
- 9:02which we're very proud of. And so those
- 9:05kind of core values people are attracted
- 9:06to. Um but when you come in work in our
- 9:09company, there are also some things that
- 9:11we don't appreciate that you do. Like
- 9:12for example, we don't welcome uh
- 9:15political discourse in our inside our
- 9:17company.
- 9:18Take it home. You guys talk about
- 9:19politics outside the company. Um we
- 9:23>> Yeah.
- 9:23>> [applause]
- 9:27>> Uh
- 9:27we we are
- 9:29the company is an apolitical company.
- 9:31You know, we're bipartisan. We want
- 9:33America to succeed.
- 9:34And and um uh we want we want uh
- 9:38whatever uh uh government is in place uh
- 9:41we'll do everything in our power to help
- 9:42America succeed. And so so that the
- 9:45discourse
- 9:46uh about about uh about race and
- 9:49religion and politics and
- 9:53all of that stuff, we tell people do it
- 9:56outside the company. It's not not for
- 9:57us.
- 9:58>> In terms of uh maybe AI regulation then
- 10:00more narrowly, um Satya was here this
- 10:02morning and what he said is, you know,
- 10:03before we talk about regulation that
- 10:06could really stymie things, why don't we
- 10:07just get some basics right? Why don't we
- 10:09get measurement right? Why don't we get
- 10:10standardization right, right? Um
- 10:13where do you land on Get engineering
- 10:14right?
- 10:15>> Get the engineering right, right.
- 10:16Translate the research in a more
- 10:17predictable way so that we're not
- 10:18fear-mongering. Keep it inside until
- 10:20we're ready to expose it. Um what do you
- 10:22think the right response is? You know,
- 10:24Demis had a proposal which was sort of
- 10:25this more FINRA-like organization.
- 10:28It's not clear what Dario wants, this
- 10:30transnational mutated thing that has
- 10:33some sort of control. Where do you land
- 10:35on this, the sort of perspective of what
- 10:37what do we need right now?
- 10:38>> You know, regulation should solve actual
- 10:41problems.
- 10:43And so the question is, what actual
- 10:44problems have we enjoyed?
- 10:45>> Right.
- 10:46>> And and um
- 10:48uh if you look at look at the actual
- 10:50problems, um all of the actual problems
- 10:52so far have come from the labs.
- 10:55And the reason for that The reason for
- 10:57that and and and just in their defense,
- 10:59the reason for that is because they have
- 11:00the most compute.
- 11:01>> Right.
- 11:02>> And the reason for that is because
- 11:03they're trying to solve uh the frontier
- 11:05problems. And so in their defense, and
- 11:08so it's sensible that that um the labs,
- 11:13the frontier labs, will be where the
- 11:15most danger come from.
- 11:17It is unlikely that a high school
- 11:19student
- 11:20uh did something because they just
- 11:22simply won't have enough compute.
- 11:23>> Right.
- 11:24>> And so
- 11:25uh it's unlikely that a startup will be
- 11:27the reason because they won't have
- 11:28enough compute. It's it's uh they you
- 11:30know, in fact, you you look across the
- 11:32planet and everybody won't have enough
- 11:34compute with the exception of the
- 11:35frontier labs.
- 11:36And so, so now the question is if you
- 11:38look at what actually happened. Um and
- 11:41they're doing pioneering work. It's
- 11:42really very hard. Um they're
- 11:44transitioning from research to
- 11:45engineering. Um I could imagine and
- 11:48they're they're they're they're
- 11:49obviously building some of the most
- 11:51consequential technology and companies
- 11:53in the world. Uh they're building their
- 11:55company, they're building their culture,
- 11:56they're building the technology, they're
- 11:58building engineering, they're building
- 11:59products all at the same time. And so, I
- 12:01I can understand it's a little bit hair
- 12:03on fire.
- 12:04Um uh but nonetheless,
- 12:06uh the four incidents from one lab, the
- 12:09one giant incident from the other lab,
- 12:11um the first thing that you have to do
- 12:13is just root cause the problem from an
- 12:15engineering perspective. What happened?
- 12:18What could we have done differently?
- 12:20And what are we going to to implement
- 12:23and institutionalize, whether it's
- 12:24technology or methods or processes, and
- 12:27make sure that we don't let it happen
- 12:28again. Now, I would bet you money that
- 12:31in every single one of those cases is
- 12:33within their control
- 12:35in the future to prevent it.
- 12:39Because the alternative it if it's not
- 12:42in their control, and I'm sure that they
- 12:43are. I'm sure I'm sure that I'm sure
- 12:45those four four incidents won't happen
- 12:47again. Um they I'm sure they root caused
- 12:49it and fixed it. I'm sure uh they have
- 12:52now technology much better technology
- 12:54for, you know, sandboxes and runtimes
- 12:57and monitors and continuous in
- 12:59continuous monitors. And you know, and
- 13:01so, I'm I'm certain they have much much
- 13:03better technology now.
- 13:05The alternative is also
- 13:07unlikely, which is for them to say,
- 13:10"Look, we had these
- 13:11incidents.
- 13:13After we're done analyzing it, we came
- 13:15to the conclusion we don't know anything
- 13:17that happened,
- 13:18and we have no idea how to control it,
- 13:21and we're asking society for help."
- 13:23>> Yeah.
- 13:23>> Now, if that's the case, then we ought
- 13:25to, you know, a bunch of a bunch of
- 13:27companies with engineers ought to send
- 13:29engineers in. I mean
- 13:30and we should advise them if we can, but
- 13:32I doubt it. I think they they have
- 13:34extraordinary people they got this
- 13:35handled.
- 13:35>> are we're not operating in a vacuum.
- 13:37[clears throat] David, last night you
- 13:38informed me that there is a Chinese lab,
- 13:40the makers of GLM, who are going to put
- 13:423 billion towards a recursive
- 13:44self-improvement run. So, maybe you
- 13:46could tee that up for Jensen.
- 13:48>> what was announced. Yeah, zhipu.com the
- 13:50founder just raised 5 billion and said
- 13:52that one of their priorities going to be
- 13:54trying to get
- 13:55recursive, you know,
- 13:56AI that trains the next AI and to try
- 14:00and automate as much of that as
- 14:01possible.
- 14:03Yeah, I think that I mean
- 14:04>> Well, this is the new sexy phrase, but
- 14:06as you guys [snorts] know, RSI is a
- 14:09combination of a system of ideas. It's
- 14:13it starts everything with
- 14:15in context stuff. It starts with skills.
- 14:17It starts with reflection. It starts
- 14:19with, you know, reinforcement learning
- 14:21and synthetic data generation. And these
- 14:23are all very sensible ideas that causes
- 14:26AI to get better at solving a problem,
- 14:30you know, over time. And you could also
- 14:32have low rank
- 14:35All of that stuff doesn't include the
- 14:36weights. You could actually improve the
- 14:38weights and it's called Laura.
- 14:40Laura could be could be improved in
- 14:42synthetic data generation, reinforcement
- 14:45learning, enhance it without training
- 14:47the the base model itself. And then over
- 14:49time, you could train the base model
- 14:51again with all of that experience. And
- 14:53so I I think I think it's a sensible
- 14:55thing that that you're going to use the
- 14:58technology
- 15:00to enhance productivity of all kinds of
- 15:02tasks including building AI. I think
- 15:05that's a very logical idea. And and I'm
- 15:07I'm certain that everybody is using it
- 15:09in some degree. It's just this phrase is
- 15:12now being used
- 15:14to weaponize the technology in some way
- 15:17and maybe to turn the the technology
- 15:19>> going to spiral out out control is the
- 15:22impression they're trying to give, but
- 15:23you don't believe that's real.
- 15:25>> No, no, of course not. And the reason
- 15:26for that is because you could RSI all
- 15:28day long inside your company, but when
- 15:31you release a product, you've got to
- 15:33evaluate it, don't you? You have to test
- 15:34it again, don't you? You have to make
- 15:36sure that there's no regression, right?
- 15:38And so the basic process of control
- 15:43These labs are going to as they move
- 15:44from labs to engineering, they will have
- 15:46much, much better control.
- 15:48>> Right.
- 15:49>> And when they have much better control,
- 15:50and that that control comes from methods
- 15:52and knowledge and practice and tools and
- 15:55technology. All of those things that
- 15:56leads to better control, verification
- 15:59and eval.
- 16:00It's going to enable RSI to be done
- 16:03inside the company and for good products
- 16:04to be released outside the company.
- 16:06>> open source for a second. I mean, this
- 16:07hugging face
- 16:09we we were communicating about this and
- 16:10I said it's going to be one of the most
- 16:11consequential
- 16:13um
- 16:15acquisitions. I don't even want to call
- 16:16it a transaction cuz I think it's more
- 16:18important than that.
- 16:20Give us your first principles
- 16:22explanation of open source versus closed
- 16:24source versus open weights and how the
- 16:26ecosystem should fit together over time.
- 16:28>> The world needs both closed models and
- 16:30open models. Um you want you want to use
- 16:33I use as much closed models as I can.
- 16:35This weekend I I used four of them. And
- 16:38and [clears throat] they work
- 16:38terrifically. They're frontier, they're
- 16:40great experience. They right, they work
- 16:42incredibly well. They're getting better
- 16:43all the time.
- 16:44Uh and and the way I think about closed
- 16:46closed closed models is kind of like
- 16:49bottled water,
- 16:50you know?
- 16:52Water is free, you guys. I don't know if
- 16:54I've told you guys, but water is free. I
- 16:56I don't want to, you know, burst
- 16:57everybody's bubble, but water is free.
- 16:59And this morning I used a lot of free
- 17:00water taking a shower. And so
- 17:03you use the right water in the right
- 17:05places. And this is no different than
- 17:07electricity. This is, you know, this is
- 17:09no different than all kinds of
- 17:10commodities that we use in the world.
- 17:12You need both. Now, in the case of open,
- 17:14the The why you you need it is because
- 17:17it could be for sovereignty reasons,
- 17:19privacy reasons,
- 17:20uh proprietary technology reasons. Look
- 17:22at the facts. The facts are in the last
- 17:256 months
- 17:27$400 billion
- 17:28of venture funding went into AI native
- 17:31companies.
- 17:3280% of them use open models. If not for
- 17:35open models, how could they build their
- 17:38dream?
- 17:39>> Right.
- 17:39>> Because their dream could be different.
- 17:41Obviously, it'll be different than the
- 17:43labs, the frontier labs dreams. I mean,
- 17:45there's America has so many different
- 17:47ways to innovate. That's one of our core
- 17:49strengths.
- 17:50Great ideas are just coming out of the
- 17:51fountain. And And so, open models
- 17:54enables that. Open models enables every
- 17:56single If we want to win the AI race,
- 17:59it's not about a few technology
- 18:01companies winning the AI race. It's
- 18:03about every company in in America. Every
- 18:06comp Every company, every industry,
- 18:08every
- 18:09researcher, every
- 18:11teacher, every student, every startup,
- 18:14everybody wins. Some of them will use
- 18:17closed models. A lot of them will use
- 18:19open models. There's 10 million open
- 18:21source
- 18:22>> Does it matter if the breakdown Well,
- 18:23let me just ask, does it matter if the
- 18:24model the open models come from China or
- 18:27the US?
- 18:28>> Well, we're doing everything we can
- 18:30um to make a contribution in open
- 18:32models.
- 18:34However, the moment you download
- 18:37Like for example,
- 18:39probably the vast majority of the
- 18:41world's contribution to open source
- 18:43today is coming from China. They just
- 18:45have a lot more engineers. They produce
- 18:47everything in large scale because it's a
- 18:49larger country. And so, they produce
- 18:51science and math students in volume.
- 18:53Right. That's one of our disadvantages.
- 18:55They're manufacturing them through
- 18:57amazing universities like Tsinghua
- 18:59University in high volume. Well, they
- 19:01contribute to open source today. We
- 19:03download Linux. We download Kubernetes.
- 19:05We download all the software. A lot of
- 19:08it has been touched by Chinese.
- 19:11Once you download it, it's yours. We
- 19:13fork it, we improve it, we make it ours.
- 19:15And so, we
- 19:17when you download one of these Chinese
- 19:19models, it just happens to be made by
- 19:21some really great researchers in China,
- 19:23but it's now yours.
- 19:25>> So, what exactly
- 19:25>> Whatever you want to do with it.
- 19:26>> So, what exactly is the race?
- 19:29>> The race?
- 19:30>> Yeah.
- 19:31>> I think that's That's a really good
- 19:32point. My point is, the race is really
- 19:35about who exploits the technology best.
- 19:40You know, in the last industrial
- 19:41revolution, all of the inventors were
- 19:44Maxwell, Volta, Ampere. None of them
- 19:46were American.
- 19:48They were Right? The last industrial
- 19:50revolution came from Europe, but we
- 19:52exploited it. We took advantage of it
- 19:55socially better than anybody else in the
- 19:57world, and look how it turned out for
- 19:59us. I want to make sure that this next
- 20:01generation happens just like this.
- 20:03>> Yeah, yeah. So, why why are the
- 20:05communists getting their message out so
- 20:07successfully here right now?
- 20:11>> [laughter]
- 20:14[clears throat]
- 20:14>> You know, I I think first of all, the
- 20:16narrative is much more practical.
- 20:19The narrative is much more practical.
- 20:21Nobody in China is saying that there's
- 20:24end of this and end of that and
- 20:27you know, cataclysmic this and
- 20:30you know,
- 20:30>> Doomer that.
- 20:31>> Doomer that.
- 20:32>> They're just much more pragmatic about
- 20:33it. They see AI as a technology that's
- 20:35going to advance their economy, advance
- 20:37their society. And they don't have these
- 20:40these groups who are basically saying
- 20:41it's going to end civilization.
- 20:43>> And we're making it up. The part that is
- 20:45frustrating is if it was true, if it was
- 20:47true, then we ought to talk about it and
- 20:49go do something about it.
- 20:50>> Right.
- 20:51>> Even even if it's true,
- 20:53we ought to spend more time doing
- 20:55something about it than worrying a bunch
- 20:57of people who can't do anything about
- 20:58it. It's our job to build it right.
- 21:00>> And that's Has there ever been a point
- 21:02in history where so many people have so
- 21:05vehemently said something that is so
- 21:06untrue?
- 21:08>> And they're measurably they're they're
- 21:09actually demonstrably untrue. And it
- 21:12actually makes sense as untrue.
- 21:15It's not based on science. It's not
- 21:16based on research. Everything that's
- 21:18based on science and research proves
- 21:19otherwise.
- 21:20>> of the frontier. Humans have never been
- 21:22there. We've never seen it. Therefore,
- 21:23we're scared of it. And therefore, it's
- 21:25easy to tell everyone to be scared of
- 21:26it.
- 21:26>> it could be life experience as well,
- 21:28David. Um so let let me give you an
- 21:30example.
- 21:31When I first graduated from school, I
- 21:33was an engineer and I didn't do that
- 21:35much typing.
- 21:36And the reason for for that is because I
- 21:38was the first generation before software
- 21:40software became popular. We had to go
- 21:42build the computers to make software pop
- 21:44possible. Could you imagine
- 21:47in this generation every single engineer
- 21:49who came into the world of engineering,
- 21:51you spend all your time typing.
- 21:54Literally, that's what you do.
- 21:56When you get a job, they give you a
- 21:57laptop, they give you a chair, and you
- 21:59start typing. You you type all day long.
- 22:02You type from the moment you wake up to
- 22:03the
- 22:04Well, there was engineering before
- 22:05typing.
- 22:06>> Right.
- 22:07>> And so so can you imagine that the world
- 22:10has a amount of engineering work to do
- 22:12where most of it is not typing anymore?
- 22:15Sure.
- 22:16We were We had busy engineers before
- 22:18typing. I think we're going to do a lot
- 22:20of great engineering after typing.
- 22:22>> Yeah.
- 22:22>> When I say typing, I mean coding. I mean
- 22:25yeah. And so even at Nvidia, when
- 22:27software engineers talk to me, I I tell
- 22:29them
- 22:30you're just typing. I've been saying
- 22:32that forever, but obviously for fun. And
- 22:35I I told I tell them my favorite key is
- 22:38backspace.
- 22:40And and the reason for that is because
- 22:42the best software is the smallest
- 22:43software. Yeah. So I I want you to use
- 22:46backspace as often as
- 22:47>> actually talk about Nvidia. I Let's
- 22:50Let's do a little teardown of Nvidia. So
- 22:53teardown meaning just explain the pieces
- 22:54cuz there's a lot of strategy at play.
- 22:57Let's start at the absolute bottom. So
- 22:58>> Oh, no.
- 23:03>> This is not planned, but we know who it
- 23:04is.
- 23:05>> Oh, no.
- 23:07No.
- 23:09Mr. President.
- 23:12>> Oh, yes, sir.
- 23:14Um I got to tell you something.
- 23:16It
- 23:16I w-
- 23:17I uh if it wasn't because of you
- 23:19calling, I would I'm on stage with the
- 23:22besties. I'm on stage with the besties.
- 23:24>> [laughter]
- 23:25>> I'm on stage with the besties. I'm on
- 23:27stage with Sachs and
- 23:29>> Oh my god.
- 23:30>> You kn- you know the whole group.
- 23:33Yeah, Jason's here, Chamath's here,
- 23:35David and David's here.
- 23:37Yeah, I'm sitting in front of a few
- 23:38thousand people.
- 23:41And we're talking as it turns out we
- 23:42were we were talking about you.
- 23:45>> [laughter]
- 23:48>> Good job, sir. Good job. The fact that
- 23:50you saw through all of that, I mean,
- 23:51there's a lot of complexity and the fact
- 23:53that it matters you saw through all of
- 23:55that and and I you know, we're all just
- 23:57really grateful.
- 24:00>> Tell him I said hi.
- 24:02>> [laughter]
- 24:04>> So, you want to say
- 24:05hi to the crowd?
- 24:06>> Jason would like Jason would like to put
- 24:08you on the
- 24:08>> Even Jason.
- 24:09>> Come on, speaker speaker mode.
- 24:12>> How do you put it How do we put on
- 24:13POTUS?
- 24:15On speaker? Put him on speaker? Speaker?
- 24:16Yeah.
- 24:17Right into the microphone.
- 24:18>> Yeah, we're going to get a mic.
- 24:19>> Hang on a second, sir.
- 24:19>> Hold on, sir. We're getting a
- 24:20microphone.
- 24:21>> Mr. President.
- 24:22>> Yes.
- 24:23>> You're now talking to the planet.
- 24:25>> See, the great thing about life is that
- 24:28Jensen can develop the most complex
- 24:30computer chip in the world that nobody
- 24:32can copy for 10 years, but he can't
- 24:34figure out to put me ON SPEAKER. THAT'S
- 24:36>> [laughter]
- 24:39>> THIS THING, WE HAVE TO REMEMBER THIS
- 24:40ONE.
- 24:41SO,
- 24:42>> ROASTED.
- 24:43>> the AI, it's almost a conspiracy.
- 24:46And the happiest group is China, and
- 24:49China's very happy. And I can even say
- 24:51in the country, a lot of states are
- 24:54happy that weren't going to get anything
- 24:56because they're being
- 24:57uh inundated by people that want to be
- 24:59there. But now all of a sudden you see
- 25:01they're building in Finland. They want
- 25:02to build one Google wants to build a big
- 25:04one in Finland, which I'm not happy
- 25:06about because they were unable to get
- 25:08permitting. And I'm telling you it's all
- 25:10a hoax. The data centers are great and
- 25:13they make people wealthy and they make
- 25:15states wealthy and it's the oil of the
- 25:17next 20 25 years. It's bigger than the
- 25:19internet and the AI,
- 25:21you know, much more so.
- 25:23And they're just playing right into the
- 25:26hands of a lot of people that don't want
- 25:28to see it happen and that could be
- 25:30political people and could also be
- 25:32China.
- 25:33And we're not going to let that happen.
- 25:34It's a it's a hoax. And
- 25:37>> You're right. We're not going to let
- 25:38that happen, sir.
- 25:39>> No, we're not going to let it happen.
- 25:41The The robots are not going to be
- 25:43taking over the world and that's not
- 25:46going to happen. You know, my uncle was
- 25:48a the top probably maybe the best of all
- 25:51time frankly but professors at MIT for
- 25:5441 42 years
- 25:57and can known as being one of the most
- 25:59brilliant men and he was he was there
- 26:02for 41 years as
- 26:04the top. He was like at the top top of
- 26:06the ladder top of did many things.
- 26:08Jensen knows all about it but did many
- 26:10things. So I have a little genetic
- 26:13a little genetic strength if you believe
- 26:15in the resource [laughter] theory but I
- 26:16do. I have genetics I
- 26:18>> That That explains why you know so much
- 26:20about AI.
- 26:22>> Well, I know about AI. I know I also
- 26:24have common sense about AI.
- 26:26The robots will not be taking over
- 26:28the AI will not be taking over the rest
- 26:31of the world. The whole thing is a hoax.
- 26:33Now with that we have to be a little bit
- 26:35careful. We have to very be you know, we
- 26:38have to do things and we have to do them
- 26:40prudently but that doesn't mean we're
- 26:42going to stop an industry because you
- 26:44know, as we work on the next 10 years
- 26:46about how to destroy it. So I'm with you
- 26:48all the way. I didn't even know how you
- 26:50felt about it. I assumed you felt the
- 26:51same way as me.
- 26:52>> Yes, sir.
- 26:53>> And we If we're going to lead, and I
- 26:55have an expression, it's whoever wins AI
- 26:57wins. That's how big it is. It's bigger
- 26:59than the internet. And whoever wins AI
- 27:02wins, and we can't let this kind of
- 27:03stuff happen. And that includes very
- 27:06much includes data centers. There are
- 27:07communities that were dying that have
- 27:10data centers right now, and now they're
- 27:11wealthy communities.
- 27:13>> Yeah.
- 27:13>> Really wealthy communities.
- 27:14>> We're We're going to make sure that
- 27:16We're going to make sure that everybody
- 27:17We're going to make sure that everybody
- 27:19wins in the AI race in America. Every
- 27:21industry, every company, every state,
- 27:23every people.
- 27:25>> Good. Well, I feel strongly about it,
- 27:27and I have the position that can do
- 27:28something about it. We're not going to
- 27:29let that stuff happen. So, I have no
- 27:32idea who's at the meeting. I have no
- 27:33idea who the hell I'm talking to, but
- 27:35I'll say
- 27:35>> [laughter]
- 27:38>> Did you Did you hear that? Did you hear
- 27:40that? Thousands of people are clapping
- 27:43for you, sir.
- 27:44>> All I know if you're there
- 27:46>> [applause]
- 27:47>> they'll listen to Jensen, but uh he's
- 27:50done an amazing job, and David has done
- 27:52an amazing job, and and good luck to
- 27:54everybody, and uh
- 27:56we're going to stay with the future. The
- 27:57country has never done better. We have
- 27:59$20 trillion of investment coming into
- 28:01the country, and that's as opposed to
- 28:04much less than $1 trillion under Sleepy
- 28:07Joe Biden. And that was 4 [laughter]
- 28:08years. This is in 1 year.
- 28:11So, you know, it's It's really the
- 28:12country has has never The country has
- 28:14never seen anything like it, and we're
- 28:16going to keep it going. And so, thank
- 28:18you all very much.
- 28:19>> Thank you, Mr.
- 28:20>> Thank you, Mr. President.
- 28:21>> Okay. I'll call you back later. Thank
- 28:24you, Mr. President. Thank you.
- 28:26Um
- 28:27I
- 28:27>> Well, that was unique.
- 28:28>> I thought it was a bit
- 28:30>> Did you know [laughter] that was
- 28:30happening?
- 28:31>> I thought it was a bit.
- 28:32>> You thought it was a bit?
- 28:33>> You thought it was a bit?
- 28:34>> No, it was real.
- 28:35>> I thought it was a bit at first, and I
- 28:36was like, put him on speakerphone.
- 28:38>> [laughter]
- 28:38>> Wow.
- 28:40And he calls [clears throat] you
- 28:41>> How do you How do you think
- 28:42>> He calls you any hour of the night,
- 28:43right?
- 28:43>> Well, we we were we were in the uh we
- 28:45were in the Oval that time when he
- 28:47called you.
- 28:47>> I was I was
- 28:48>> was I was in the Oval.
- 28:49>> You were asleep and he like said, "Wake
- 28:50him up." And he I felt I felt so bad
- 28:53because he's like, "Who's coming to this
- 28:54dinner?" And we go through the list and
- 28:55he's like, "Well, what about Jensen?" I
- 28:56said, "No, sir. We I he's on vacation
- 28:58cuz he had to postpone this vacation for
- 29:005 years."
- 29:02And he's like, "Get him on the phone."
- 29:03>> [laughter]
- 29:04>> What's vacation?
- 29:05>> But what do you Why do you think he sees
- 29:07through the hoax? It's It's This is the
- 29:08thing I was asking you
- 29:09>> quite an extraordinary thing.
- 29:10>> It was It's polling minus 80.
- 29:12>> Yeah.
- 29:13>> So, for anyone else that's sitting in
- 29:14the Oval Office, you're going to do
- 29:16what's popular. You're representing the
- 29:18people. This is what everyone wants.
- 29:20They want to shut down the data centers,
- 29:21end AI. It seems to be the popular thing
- 29:24in the moment. But he says it's a hoax,
- 29:26and he calls it. How does he do that?
- 29:29>> I got to tell you, I'm not sure. And the
- 29:30reason for that is because a lot of
- 29:31people are falling for it. And so, the
- 29:33fact of the matter is it's complicated.
- 29:35You know, at first, I mean, if you look
- 29:37at the story if you look at the stories,
- 29:39it's all anchored on two things. The
- 29:41first thing that it was anchored on was
- 29:42national security. And recently, that
- 29:44was all blown blown to bits.
- 29:46>> Right.
- 29:46>> And so, no no that story is no longer
- 29:48anchored on national security. Now it's
- 29:50anchored on safety. Now,
- 29:52if you want AI to be safe, um the first
- 29:55thing is we need to make sure that the
- 29:57the labs that are building it are in
- 29:59control. That there there good tests for
- 30:01them. Uh if we would like to have third
- 30:03parties uh
- 30:05uh to to um uh make sure that a third
- 30:08party evaluate third party evaluators
- 30:10are available. That's no different than
- 30:11financial control. You guys know we have
- 30:13auditors.
- 30:14And the auditors are quite quite um they
- 30:17don't have to be as expert as we are in
- 30:19our business, but they just have to ask
- 30:21the right questions. And um I I think I
- 30:23heard somebody say that it's good to
- 30:24have uh independent auditors or
- 30:27evaluators, but they just have to have
- 30:29multiple. I agree with that, too. Just
- 30:31as there's multiple evaluated and
- 30:33auditors, it makes sure that one company
- 30:35doesn't become, you know, pill or
- 30:37somehow influenced um for for whatever
- 30:40reason. And so, you you know, there's a
- 30:41lot of different ways that you could
- 30:42solve this. Um and so, that I think the
- 30:44number one thing is let's build the
- 30:46technology safely.
- 30:49Let's make sure that the testing of it
- 30:51is safe. And I I recognize completely
- 30:54that that what what is being built is
- 30:56extraordinary. Um but, these are
- 30:57extraordinary companies and and we ought
- 30:59to hold them to to uh extraordinary
- 31:01standards. Um and that they want to be.
- 31:03And they want to be that
- 31:04>> I wanted to go back to open source for a
- 31:06second. [clears throat]
- 31:07Um
- 31:08a year ago,
- 31:09we were taking it very seriously. It was
- 31:122 years, 18 months behind.
- 31:13>> The one thing that, you know, one one of
- 31:15the as you guys know, one of the
- 31:16challenges
- 31:17when you're on the call with President
- 31:19Trump, it's hard to
- 31:20say something.
- 31:22Um
- 31:23>> [laughter]
- 31:24>> I'm going to get in trouble for that,
- 31:25I'm sure. He's going to call me up up on
- 31:26that. But, anyhow,
- 31:28I what I was going to tell him and and
- 31:30and all of you is that AI is creating
- 31:32enormous number of jobs. The the the the
- 31:35thing that he wanted more than anything
- 31:36at the beginning of the the
- 31:37administration and that my first phone
- 31:39call with him, my first time I met him,
- 31:41is that he wants to create jobs in
- 31:44America. He wants to re-industrialize
- 31:46the United States. He wants to make sure
- 31:48that United States has the energy to
- 31:49support the next industrial revolution.
- 31:52Without energy, there's no industrial
- 31:53growth. And so, he wants to make sure
- 31:56that there's energy growth, that there's
- 31:58job growth, that the re-indust- we're
- 32:00re-industrializing the supply chain.
- 32:02Look at everything that we're doing
- 32:03right now. All of it is happening right
- 32:05now as we speak. We're creating more
- 32:06jobs than ever. We're creating software
- 32:08jobs. Uh we were just talking about
- 32:10earlier, 400 billion dollars of venture
- 32:13financing went into the AI industry just
- 32:16recently. 6 months. Well, that's created
- 32:19a ton of jobs. That's created a ton of
- 32:21jobs. Um it's created, you know,
- 32:23obviously enormous amount of demand for
- 32:25compute, which I'm which I'm I'm happy
- 32:27about. Um which is also which is also
- 32:29creating a lot of demand for data
- 32:31centers. And we had to talk about that.
- 32:32I think I was just I was talking to um
- 32:35uh Governor Abbott uh
- 32:37uh of uh Texas, and he was he was uh he
- 32:40wants to appeal to the industry to make
- 32:41sure that we are we are empathetic to
- 32:44the small communities as we're building
- 32:46data centers all over all across
- 32:48America, just to be better listeners.
- 32:50>> Let's actually talk about that for a
- 32:51second. Let's [clears throat]
- 32:53What's incredible about Nvidia, if you
- 32:54if you break down the component parts is
- 32:57you've effectively had to become the
- 33:00bank of AI to get the ecosystem going.
- 33:03And you've had to do it at all levels.
- 33:05You know, you just did this thing with
- 33:06Clover Leaf where you're doing Lam Power
- 33:07Shell.
- 33:08You did this great thing with BlackRock
- 33:11and Goldman and all these folks to to
- 33:13essentially create the financing
- 33:14capability. Walk us through your capital
- 33:17allocation strategy. Like, what has to
- 33:19happen
- 33:20to get a broader ecosystem folks to be
- 33:23able to come in and underwrite this next
- 33:24phase?
- 33:25>> Well, we're we're creating As you guys
- 33:27know, this is a new industrial
- 33:28revolution. And and um every aspect of
- 33:31it is true. Um this new industry
- 33:34requires manufacturing just as just as
- 33:37uh the the the the um electricity,
- 33:39internet, and now AI.
- 33:41We power anything.
- 33:43We can
- 33:45find anything. Now with AI, we can ask
- 33:49and know anything. Isn't that right? And
- 33:51so that's our future. We're tapping to
- 33:52the ether and we can ask it of anything
- 33:54we want and it could explain it to us.
- 33:56Now, in order for that to happen, it's
- 33:58got to produce the intelligence. And so
- 34:00that's a production process, which is
- 34:02the reason why this infrastructure has
- 34:03to get built. But once you get the
- 34:05infrastructure built, the question is um
- 34:07what about all of the other layers
- 34:09across the United States?
- 34:10Uh this industry isn't just about the
- 34:13model. It's not just about the chips.
- 34:15It's mostly about the applications on
- 34:18top. It's mostly about the
- 34:20infrastructure layer, the data centers,
- 34:22and all the infrastructure, the the the
- 34:24the construction, the electricity, the
- 34:26power generation. That all of that is
- 34:29involved. And so I look across the
- 34:31entire ecosystem and look for
- 34:32bottlenecks. And if there are places
- 34:34where extraordinary companies are being
- 34:36built
- 34:37>> Constraints.
- 34:38>> Constraints. Extraordinary companies
- 34:39being built. Maybe it's
- 34:43supply chain that has to get scaled up
- 34:46so that when we're ready to deploy
- 34:49compute that they'll be ready for us.
- 34:50Land power shell. And so, this is no
- 34:52different than looking at the supply
- 34:54chain upstream. You know, I I probably
- 34:57think about the long-term supply chain
- 35:00more than most because our company's
- 35:01really large and and in order for us to
- 35:04succeed, a whole bunch of companies has
- 35:06to support me. You know, it's got to
- 35:08Corning has to you know, Wendel
- 35:11Corning has to support me, Lumentum and
- 35:14you know, TSMC of course and memory
- 35:16companies and and so, we started working
- 35:18with all of these companies long before
- 35:20the revolution
- 35:21the growth came so that the growth could
- 35:24happen. Now, I'm going to Now, I'm doing
- 35:26a downstream.
- 35:26>> I I have a cycle tends to be that that
- 35:28the earnings over time over long
- 35:30stretches of time tends to move up the
- 35:32stack right towards the application
- 35:34layer where you can over earn for larger
- 35:36periods of time.
- 35:37Um
- 35:38I mean, you bought Hugging Face. Now,
- 35:40you're sort of in the actively in the
- 35:42serving business. I mean, it seems
- 35:44pretty natural that
- 35:46products like Open Router make a lot of
- 35:48sense. It seems pretty
- 35:50obvious that, you know, there are better
- 35:51versions of ways to build things like
- 35:53Bedrock. I'm sure you think about it.
- 35:55What's the natural conclusion because it
- 35:57seems like the folks up here have no
- 35:59issue trying to move down.
- 36:01>> Mhm.
- 36:02>> And you have the best balance sheet,
- 36:03these incredible engineers, and you have
- 36:05the proven experience to make it right
- 36:08and engineer the product and get it out.
- 36:10So, how do you think about looking up
- 36:12and saying, I could probably do that?
- 36:14>> The The reason why Nvidia runs every
- 36:16single model in the world we we're the
- 36:19only it's incredible. Last year, about a
- 36:21year and a half ago, the only thing we
- 36:23ran was Open AI.
- 36:24>> Yeah.
- 36:25>> And now look at amazing models are
- 36:27available. The Meta Muse is available.
- 36:29You got Grok is available. Grok bot's
- 36:31incredible.
- 36:32Um we now run Gemini uh and uh Anthropic
- 36:35is is uh scaling up on our platform as
- 36:37well. Uh since a year and a half ago,
- 36:40you got all these frontier AI models
- 36:41that are now open, that are available.
- 36:44So, the number of models that are are
- 36:45are growing. Um there's a whole bunch of
- 36:47companies that I won't mention that are
- 36:49building uh frontier models as well.
- 36:51And the the number of AI labs are
- 36:53growing. Yeah, the the Ineffables, the
- 36:57uh the Reflections, the right? The the
- 37:00list goes on. The Physical Intelligence,
- 37:01the list goes on. Okay? And so, all of
- 37:04these labs are building on Nvidia. And
- 37:06the reason for that is because
- 37:08as a company, I'd rather for us to help
- 37:11everybody succeed instead of taking a
- 37:14slice out.
- 37:16And so, we would go up as far as we need
- 37:19to,
- 37:20but as low as possible.
- 37:22>> Huh.
- 37:22>> Our strategy is go up as far as we need
- 37:25to and as low as possible. And the
- 37:26reason for that is because if I do that,
- 37:29if I solved if it if not for Nvidia
- 37:31creating cuDNN, all of the frameworks
- 37:34wouldn't exist. If not for us creating
- 37:36Megatron uh make Megatron core, uh then
- 37:40all of the large-scale training wouldn't
- 37:41have happened.
- 37:42>> Training wouldn't exist.
- 37:42>> Um so, we we go and we invent all the
- 37:44technology necessary as far as we need
- 37:46to and then we let a thousand flowers
- 37:49bloom. And so, that posture allows us to
- 37:52be, quite frankly, the only
- 37:54>> be honest. I I agree with you.
- 37:57The pushback would be that
- 37:59it really would be great to have more
- 38:00competition at the hyperscale layer.
- 38:03And I think you've done a great job
- 38:04supporting the neo clouds. There's some
- 38:06I'm by the way, I think you introduced
- 38:07me to Nimbix. Superb. Great. Everything
- 38:09there amazing. But we need like 50 of
- 38:11these guys. We need a hundred of them.
- 38:13We need a thousand of them. And they
- 38:15just may take
- 38:16some
- 38:17>> Yeah. Yeah, you know, Chamath, it's just
- 38:20I'm surprisingly uncompetitive.
- 38:23>> Really?
- 38:24>> Yeah.
- 38:25>> [laughter]
- 38:25>> That's not my thing. You know, my thing
- 38:27is kind of like for example, I'd be more
- 38:30than happy with five hyperscalers.
- 38:32However,
- 38:33um the reason noticed the early
- 38:36customers of all the neo clouds, all the
- 38:38what we call NCPs, all the early
- 38:40customers were the hyperscalers.
- 38:42>> Exactly.
- 38:42>> And the reason for that is because the
- 38:44hyperscalers plan once a year,
- 38:47but the market dynamics is so volatile
- 38:49right now that they're always almost
- 38:51wrong.
- 38:52And so with all these regional clouds
- 38:56who are agile and they can move fast, um
- 38:59they know their state or they know their
- 39:00country, they know their region, they're
- 39:02securing land, power, and shell in a way
- 39:05that's hard for somebody who sits in
- 39:07Seattle or sits in Palo Alto to be able
- 39:09to see the planet.
- 39:10>> Mhm, right.
- 39:10>> And so we now have basically a
- 39:12large-scale distributed network of
- 39:14companies that are building, securing
- 39:16land, power, shell for us.
- 39:18And [snorts] um
- 39:19and and now countries realize it's
- 39:21strategic.
- 39:22>> Yeah.
- 39:22>> So many countries are saying, "I'm going
- 39:24to take my power and only give it to my
- 39:26own companies."
- 39:27>> Right.
- 39:28>> Well, Nvidia's in that country as well
- 39:30and we could help the neo clouds in that
- 39:32country grow and and so whether it's
- 39:34whether it's Fermus in Australia, we
- 39:36just did a whole bunch of stuff in
- 39:37Australia, um brought on two more
- 39:39gigabytes, uh Southeast Asia of course,
- 39:42uh IOH and others bring on a few
- 39:45gigabytes. And so we're building a
- 39:46gigawatts, so we're building, you know,
- 39:48we're
- 39:49we're scaling up, you know,
- 39:51>> We're getting there though. I just want
- 39:52to get this one thing in. It's pretty
- 39:53clear that you're going pretty high up
- 39:56and getting very focused on open source.
- 39:58Obviously, you have your Neumotrons
- 40:00doing exceptionally well. I use them
- 40:02often. Hugging Face, Poolside, and
- 40:04Laguna, uh very very solid product that
- 40:07you're now uh aqua-hiring, hiring,
- 40:09whatever it is.
- 40:10Um and then you have your open source
- 40:12stack for self-driving, also uh very
- 40:15disruptive.
- 40:16>> So we are the frontier model in five
- 40:18domains.
- 40:19>> Yeah.
- 40:19>> And so,
- 40:21you don't seem to build products to get
- 40:23the silver medal. You seem to go for the
- 40:25gold. So, are you going for the gold,
- 40:28and will you have the best hands-down
- 40:30open-source model?
- 40:32And then,
- 40:33part B to that is, can open source catch
- 40:36up to frontier models, and are you the
- 40:38person to do it?
- 40:39>> So, the logic the logic, Jason, is that
- 40:41that we will build it because one,
- 40:45uh we can. We have the skills to do it.
- 40:48And because our customers need us to do
- 40:50it.
- 40:51>> Right.
- 40:52>> So, for example, Alpaca Myo is the
- 40:54world's first thinking self-driving car.
- 40:57And by by thinking, by reasoning, you
- 40:59don't need as much data as
- 41:01you know, you don't have to train on a
- 41:02few billion
- 41:04hours of road data. Because you can
- 41:06reason about it. Break down the problem
- 41:08into oh, I've seen this before. It's not
- 41:09exactly the same, but it's largely the
- 41:11same as that. Okay? And so, so Alpaca
- 41:14Myo, why is it necessary? Well, there's
- 41:15a whole bunch of car companies. Every
- 41:17car in the world is going to be
- 41:18autonomous. But beyond that,
- 41:21every
- 41:22ag tech, every truck, every van, and
- 41:25most of them aren't big enough in scale
- 41:27to be able to build that whole stack.
- 41:29So, I'll build an extraordinary stack
- 41:31for them. They do last mile adapting for
- 41:33their application. Now, everything that
- 41:36moves in the future could be autonomous.
- 41:38If not for us building
- 41:40some of the some of the biology models,
- 41:42the world wouldn't have it. The the ESM2
- 41:45protein found language model, we created
- 41:47that. ESM fold, open fold, AlphaFold2.
- 41:51All this stuff with with co-equivariant.
- 41:53All of that stuff technology wouldn't
- 41:55have existed if we didn't build it.
- 41:58One of my favorites,
- 41:59protein complexa,
- 42:01you know, synthesizing next-generation
- 42:03proteins and its binding. It's
- 42:05groundbreaking stuff. We built that. And
- 42:07so, we'll build that because a Lilly
- 42:08needs it, and and you know, Merck needs
- 42:11it, and others need it, and they don't
- 42:12have the capability to do it or they
- 42:14they're not yet there. And so we can
- 42:15make a real contribution. So I do
- 42:17everything out of need. I'm not trying
- 42:19to disrupt. I mean, we don't wake up in
- 42:21the morning try to disrupt anybody. We
- 42:24just wake up in the morning try to help
- 42:25everybody.
- 42:25>> Jensen, what about what about
- 42:27competitive threats that might be
- 42:28emerging to your core business? Can you
- 42:30just comment
- 42:31>> so nice.
- 42:32>> Yes. Well, I know this is well, I
- 42:34actually want to I want to just get your
- 42:36let's just call it a take. What's your
- 42:38take on Terafab 100 million square foot
- 42:40facility Elon's announced? And
- 42:43>> If anybody could do it, he can. And the
- 42:45two of us were on a flight together to a
- 42:48country.
- 42:49And um
- 42:51>> With a person
- 42:54who sometimes calls you on the phone.
- 42:56>> It was a It was a nice plane.
- 42:57And and
- 42:59And we had like, you know,
- 43:01and
- 43:02you know, Elon likes to talk about these
- 43:03things and and so uh we spent a lot of
- 43:06time talking about it.
- 43:08I I
- 43:08>> Is that
- 43:08>> If anybody could do it, he could do it.
- 43:10>> Cuz I mean, you could you design chips,
- 43:12you don't fab them.
- 43:13Could your chips be fab there or is it
- 43:15>> Well, we know we have we know a lot
- 43:17about process technology because we're
- 43:18pushing the limits of everything.
- 43:20>> Right.
- 43:20>> And you know, [clears throat] because we
- 43:22scale at such large scale,
- 43:24uh we have incredible memory technology
- 43:26inside the company. We're the world's
- 43:27best 30s company. You know, we got lots
- 43:29of amazing tech.
- 43:30>> So your take is you've talked a lot
- 43:31about it.
- 43:32>> So we could just Yeah, we could talk
- 43:33about it and and um you can't discourage
- 43:36Elon from doing it, which is one of his
- 43:38incredible That's his superpower. And
- 43:40once he decides to go do something, it's
- 43:42hard to stop him. And so I
- 43:43>> Can can you give us your take on where
- 43:44China is with advanced lithography
- 43:46systems
- 43:47um native grown?
- 43:48>> They're going to get there by 2030.
- 43:50>> By 2030? [clears throat]
- 43:51>> Yeah. And 2030 is just around the
- 43:53corner.
- 43:53>> Yeah. Also, that's Sorry, how long will
- 43:55we'll be dead at that time? So
- 43:58And does And And for China, does that
- 43:59mean the switch is flipped and then
- 44:01that's all going to go into um
- 44:03mainland fabs?
- 44:05Almost immediately.
- 44:06>> You know, the the way to think about
- 44:07China is they're
- 44:08good at high volume production.
- 44:12And this is just a matter of time.
- 44:15>> Yeah.
- 44:16>> And I you know, I I think in
- 44:18I I think in decades as well. You know,
- 44:21I've been around a long time and
- 44:23you know, for Nvidia I got to think
- 44:25about what happens next decade and
- 44:26decade after that. So, two or three
- 44:28years is is just a click. It's nothing.
- 44:31And so, as far as they're concerned,
- 44:32they're already there.
- 44:33>> They're already there.
- 44:34>> Yeah.
- 44:34>> Jason, Elon and Gwynne.
- 44:37>> We've got to run.
- 44:39America, we got to run.
- 44:40>> Yeah.
- 44:40>> Speed run. We got to speed run. Slowing
- 44:43down is definitely the wrong strategy.
- 44:45>> Well, I mean
- 44:47it [clears throat] it feels apparent I
- 44:48think to most of us in the industry that
- 44:50we're kind of in the AGI moment and it's
- 44:53a definition obviously, as smart as any
- 44:55other human.
- 44:55>> we're already there.
- 44:56>> We're there, right? And so, then super
- 44:58intelligence is the next waypoint. Based
- 45:01on what you see, based on your customer
- 45:02base, based on your history here.
- 45:04>> And Jason, I think we're there, too.
- 45:05>> You think we're at super intelligence?
- 45:07>> When you when you when you take a narrow
- 45:09segment,
- 45:11a narrow segment, I mean my my
- 45:13self-driving car, I don't want you to
- 45:14make me an omelet. I just want you to
- 45:16drive the car.
- 45:17>> Right.
- 45:17>> That is super intelligence.
- 45:19>> It's better it's better than a human.
- 45:20>> Yeah, yeah.
- 45:21>> Yeah. Like 1/10 the the accident rate.
- 45:23>> Exactly.
- 45:24Uh synthesizing proteins, you know, uh
- 45:27doing virtual screening of proteins.
- 45:29We're already there.
- 45:30>> Yeah.
- 45:30>> Yeah, yeah.
- 45:31>> Are you having fun being on the frontier
- 45:32of humanity?
- 45:36>> I like it.
- 45:37>> Yeah.
- 45:37>> [laughter]
- 45:40>> Ladies and gentlemen,
- 45:41ladies and
- 45:42>> I like it. I like it. And then guys,
- 45:44guys, it's it's it's great there.
- 45:47The future is great and we want to get
- 45:49there. The listen,
- 45:51>> Ride the wave.
- 45:51>> a lot of us don't have to work, but I
- 45:54got to tell you, it's too good not to
- 45:56be.
- 45:56>> So fun.
- 45:56>> Right? And so, so I want every we I want
- 45:59to be there. I want all of you guys
- 46:01there with me. We're all going to be
- 46:03there. We're going to be enormously
- 46:05successful together as humanity. And and
- 46:08in the meantime,
- 46:10we got to encourage them, urge them on.
- 46:12They're doing really, really important
- 46:13work as you guys know. And I want them
- 46:15to succeed. I also would love for us to
- 46:19tone down the the the drama. And and
- 46:22most importantly, we need all of America
- 46:25to come with us. That's how we make it.
- 46:27>> Ladies and gentlemen, Jensen Huang.
- 46:30>> [applause]
- 46:31>> Thanks, [music] man.
- 46:32Appreciate you. Thank you.
- 46:35>> Thank you.
- 46:35>> Thank you. Thank you.
- 46:36>> That was great.
- 46:36>> awesome.
- 46:37>> [music]
- 46:38>> Following your party.
- 46:39>> awesome.
- 46:39>> That was great.
- 46:41>> Thanks, guys.
- 46:42>> Thank you.
- 46:43That was great, huh?
About this transcript
This page contains the full transcript of Jensen Huang: The Doomer Hoax, Superintelligence is Here, and The Future of AI (ft. President Trump) by All-In Podcast, generated from the public captions YouTube serves with the video. The transcript has 9,096 words across 1,503 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.
What you can do with it
Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to choose a caption track available on the video.
Free YouTube transcript tool
YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, caption language selection, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.