Money is like oxygen. If you deny oxygen/money to the Chinese chip market, they can't grow as fast. By supplying China with chips, all the money will go back to the US and not enough money will go to Chinese chip manufacturers to help them advance the way Nvidia can.
This is what Jensen was saying and he is 100% correct here. The chip ban to China will really help them catch up because the money will recirculate since they have no choice but to buy Chinese chips, and in the end it will make them much stronger.
The fundamental contradiction of the AI maximalists like Huang is claiming that AI does not need regulated while simultaneously supporting chip exports to China. The supporting claim for the former is that we need to compete on AI with China and move as fast as possible and every day matters, while the latter implies that slowing down their compute is inconsequential.
If this principle were true, then autarchies like North Korea would be the most advantaged since they never have to worry about international competition in their domestic market.
Your confidence is not backed by any data or observation, it's just an unfounded inference.
OT but one thing I didn't realize until reading this comment:
Those used CMP 170HX 8GB were advertised in front of crypto server rigs. Those were probably spread across South Asian scam malls(shopping mall buildings built in literal sovereignty contested areas of non-China/Japan/Korea/Russia non-India non-African non-Muslim Asia, under guises of Chinese tourist destinations, that are literally rented out to local Chinese-speaking gangs engaging in all sorts of gang crimes from tech support scams to human and organ trafficking, straight out of the fictitious Chiba in Neuromancer), and they would happily sell or rent compute to Chinese AI firms if only those were more valuable than 8GB VRAM they have, which they all had all along.
GPU export to China must have become significantly less meaningful since the 170HX unlock became a thing. That alone must have provided petaflops of unsanctioned untraced junks for parts priced compute to Chinese AI startups.
There are. A lot of Nvidia chips are still stuck in Hong Kong.
What the government does is not always what corporations follow.
https://nationaltechnology.co.uk/Nvidia_GPUs_Sitting_Unused_...
https://www.tomshardware.com/tech-industry/u-s-commerce-secr...
"About 600,000 of Nvidia's advanced artificial intelligence chips are stored in Hong Kong awaiting shipment to mainland China, two people familiar with the matter said, as Beijing withholds approval for large-scale imports despite demand from tech firms" [1].
[1] https://www.reuters.com/world/china/nvidia-chips-pile-up-hon...
I do not know full details, but China did exactly that. There were articles about limiting Nvidia chip imports more than a year ago. It really highlighted the absurdity of the export ban.
To my non-expert knowledge, the reasoning behind the export ban has never been explained beyond a vague assertion that China would do some sort of evil with AI if they have bigger GPUs than some arbitrary number of transistors, and also some silliness where paying a 15% tax made some of the GPUs safe to export. It has been pretty great for Huawei.
Both China’s geopolitical situation and technological history point towards them cutting any reliance on a foreign tech stack as soon as they can.
As I see it, giving them chips just means they have less of a gap to close when their own EUV fabs come online, it can’t have any delaying effect on their plans.
Yeah, no. No one in Beijing will let Nvidia or the White House have that kind of power in the long run because the free market knows best and golly gee where will we get the money and whatnot. They will buy whatever they can to bridge the gap to a time when they own their own tap.
It's not a completely dissimilar argument to selling them the nukes or F-35s because "it will deny oxygen/money to the Chinese nukes and fighter jets market".
The best way of hampering China technologically has always been about keeping them completely dependent on foreign chip supply chain, e.g. ASML, TSMC, NVDA, ARM, AVGO, INTC, etc.
If you simply ban them from getting any western chips or chip tech you simply light a viciously hot fire underneath them to develop their own TSMC equivalent and their own chip designers, and we are now seeing them catching up rapidly. In fact, they are pretty close to catching up now, basically within one or two years...
The best reason to NOT ban exports is that it really doesn't work against a peer state, and China is close enough now to count. Given the significant economic trouble in the PRC (youth unemployment, unwinding the property bubble, overproduction of EVs) the US could definitely have played this better instead of creating a vibrant full-stack domestic AI sector in China.
This has crashed, check the markets.
They do not need a ban to light a fire under them. The fire is already lit and burning and nothing the west can do will even slightly dampen it.
You yourself claim they are within a couple of years. If that's the case you think they will stop developing if we sell them chips? That's laughable.
Keeping adversaries X-behind discourages adversaries from developing a new ecosystem since X-behind is better than building new. If an adversary with tremendous resources must recreate the ecosystem, it will be adopted by US+global companies since it will be lower margin.
Watched the interview and as far as I see he was getting 'heated' because he was speaking the obvious truth and being frustrated by poorly informed questions.
This is entirely expected from somebody like him who has a long history of being on the right side of technical argument.
He is a huge loss to Nvidia and the US - replaced by a ex-CFO.
I don't see it. Nvidia has a lot to answer for, and the fact that Jensen was very successful as a CEO does not shy away from the cracks in the structure.
It was illegal, and I'm sure if this comes to light there will be (major) repercussions, but I don't see it as an immoral move.
Cheating on your wife of 40 years OTOH...
In this case, though, I think Huang just believed these sanctions were the wrong move, and would end up long-term helping China and hurting the US. Which is fair to believe, but believing a law is counterproductive doesn't give you immunity if you break it.
This is a significant philosophical point.
The thing about that Lord Moulton said is that there are a bunch of things that are wrong that are not illegal, and we have to do that because people have to be capable of self-control. Patrick Deneen, in . . . The Failure of Liberalism said that he thinks liberalism is failing because it has come to the view that anything that is not illegal is moral.
What if he "believed fidelity was the wrong move" and that he'd end up divorced eventually anyways?
When is disloyalty and dishonesty immoral?
IMO it's not a terribly strong case to make while Boeing sells whole jetliners to China, but the law was put in place for both moral and practical reasons.
If the sanctions are to prevent buying the products of slavery, that's a moral issue, and violating the sactions is also a moral issue.
Same deal if the sanctions are to prevent selling things that will be used to carry out genocide/etc. That's a moral issue, so violating the sanctions is a moral issue.
I don't think there's an alegation that the GPUs will be used for immoral purposes, just that the US government has national security concerns. Those concerns may or may not be valid, and may or may not be served by having trade sanctions applied effectively, but I don't think that's a question of morality.
It's not de facto immoral to violate the law. It could be immoral to violate the law and conceal it.
But how do we know he wasn’t being whispered to by the smoking baddie he was trying to impress? Maybe he’d not have taken the risk were he not having this affair but we know what men do when their ego is on the line. Case in point: Musk paying people to play PoE for him and losing his shit when he was caught.
Do I think money is a good enough explanation? Yes. Could I believe that what’s in his pants also played a part? Also yes.
Since NVDA had such a flat corporate structure with Huang having ~60 direct reports, I expect that the company will be lost for a period of time as they try to get their bearings. This is an opportunity for AMD, INTC and others.
This is a $5T (well, I guess $4T now) company we are talking about. What is the board being paid for?
This is a $4T company for Pete's sake!
They had an idea which they thought will be valuable. And they worked bery hard on making that idea a reality.
People complain that companies don’t have leadership with long term vision. The leadership of NVIDIA heard the idea of GGPU computation and put the resources behind the idea to make it a practical reality. Here the gap between “nerdy idea” to “product sells like hot-cakes”, is measured in decades. That is not just luck.
Jensen Huang cheated on his wife AND conspired to evade arms control regulations, but he's still a genius.
Jensen has two core points:
- stopping selling NVidia chips to China isn't going to stop China developing AI
- stopping NVidia selling to China will ultimately ceed US leadership in chips as Nvidia's commercial edge is as much about the dominant developer ecosystem as it is the chips per se.
ie the real risk is the default world tech stack shifts to open models on Chinese chips. End result Chinese lead in both.
He has spent his entire career competing and winning and he is being told he has to stop competing in order to achieve a goal that he knows won't actually be achieved by sacrificing NVidia.
In that scenario, where somebody is suggesting you throw away your life's work for nothing - I think anybody would be animated.
As someone who isn't based in the US/EU and whose job is to deploy LLMs on premises, I can't wait for the moment when China starts selling its GPUs globally en masse with a proper CUDA equivalent. It's going to be cheaper in the long term and without all the export control drama.
It clearly didn't - don't you remember the Deepseek launch where it came as a rude shock that the much vaunted concept of a compute capacity moat was blown away?
He kept repeating that while he made the best chips, computer science was more important.
Ma Bell without the anti-trust (yet!).
Let's be clear, she is just the interim CEO (primarily because Huang neglected to ever even start a transition plan) while NVidia searches for a permanent CEO. The names I've seen floated are Rene Haas, CEO of ARM, and Ian Buck, the creator of CUDA.
You can say it was luck but it was definitely not "average".
Nvidia thrived because it paired great hardware with great software.
AMD and Intel think hardware is enough; it’s not. There’s a reason neither has ever been able to compete with Nvidia.
They have a giant ego, but I’d argue it’s probably well deserved.
Apple is/was similar. They also see that hardware without great software is useless. Spend as much; if not more on software to make your hardware shine.
Another thing is wrong was thinking that Apple should run in and build some type of AI model that’s proving to be well wrong the only thing in Apple’s way right now is memory. certainly isn’t money, and they certainly have the talent and house to fix the memory problem. The question is whether or not they have the will to do so.
No one equals Apple at this point in time when it comes to seamlessly integrating multiple devices in a useful way. Be it a desktop computer, laptop a smartphone, tablet, smartwatch, Apple Vision, AirPods with each iteration, once again in comparison to their competition, they seem to have the best hand at this time.
Memory cost is the biggest problem Apple has right now and it’s not like they don’t lack the money and with the Apple Silicon design group they certainly have the capability to fix it, or I should say design around it.
Yes, it might take 2 to 5 years to do it, but it’s not impossible, the main question is, do they have the will? And they’ve certainly shown it when it comes to processors, GPUs, modems, and other support chips that they can execute over time.
They may not execute on what you want or what I want, but in comparison to their competition, who is out there right now that can combine both the software and the hardware at a high level across the board?
So did they round something you wanted to see square, or vice versa?
Nvidia is a highly technical company we’re gonna find out how important Jensen really was to that company tech wise?
And if you look closely at their company, what have they really innovated on over the last five or six years? from a distance, like Intel, it seems like they just turn up the juice/Wattage in short, they are just cruising along.
You can criticize his recent question actions all you want, but you can’t believe he is “average”.
Due to sanctions there’s one Chinese company that seems to have become a multi-dimensional company we won’t know for sure for another two years.
Being a vertical computer company appears to be coming back in style bring on the 1980s. early 1990s.
Yeah, it pretty much does. CUDA is why they are positioned where they are. Other players like AMD and Intel could have done that, but they didn't, and if they had, they wouldn't have persevered for 10+ years with essentially nothing to show for it.
Whether CUDA will continue to maintain dominance is less clear. It may become a literal victim of its own success, as CUDA-based models gain the ability to migrate themselves to new platforms.
Western companies have had that problem. when the visionary CEO bites the dust. I think the Eastern companies are even more susceptible.
Note: Nvidia is the very definition of a one trick company I think they were in trouble long-term no operating system, Microsoft is in trouble in the other direction. No real in-house hardware ability. Long-term you have to have the ability of both in-house to adjust and iterate over the long haul with the changing conditions in tech.
He may not have 'invented' CUDA but he saw the potential and gave it resources to be created and continually developed.
How many "CUDA-like" ideas have be squashed at other companies by leaders that don't see the potential or are willing to take a multi-year loss?
IMO, it's like SpaceX vs other rocket companies - especially Blue Origin.
Bezos, at least at first, had just as much money if not more than Musk. So why did SpaceX do so much better than them and every other rocket company. I despise Musk personally, and he's generally not as smart as he makes himself out to be - but there is something there he does differently than others in his position, some with similar resources as he does as well.
Let's expect better of leadership than to be criminal.
Unless it's like super far in, if this counts as heated then I just don't know what you expect.
If the Chinese Communist Party are going through such great lengths to obtain these advanced Nvidia chips, it further invalidates a lot of what I've been reading here and elsewhere about China "catching up" or that being the leading models isn't what matters (commoditization). You don't do stuff like this unless you really need something.
Unbeknownst to them, the Russian was actually at that time trying to make contact with western intelligence, as he was keen to defect! He was just a bit too awkward to respond to flirting.
He did eventually manage to defect to Britain, and died in 2025. His name was Oleg Gordievsky.
WSJ reports that U.S. intelligence says Chinese intelligence approached Alice to become an asset after she met Jensen in March 2024. So, she was an opportunistic info source, not targeted in advance.
That said, I don't have any special insight into the specifics of who was driving the demand backing Megaspeed.
China catching up does not mean they are not still behind.
Also, China is a big country with many companies. While some may be trying to catch up via local development, others may still prefer getting western hardware right now e.g. in order to get a competitive advantage over those catching up.
If they don't need NVIDIA then why is the now former CEO of NVIDIA trying so hard to allow exprt controls to be relaxed.
And it's because brute forcing only allows you to not fall behind, but you can't catch up to frontier models or surpass.
But here's a link to the podcast episode if you want to give it a listen. It will inform a lot to the discussion taking place here and help you understand a few of the things he said that are probably now wrong: https://www.dwarkesh.com/p/jensen-huang
One other little crack in the dike. Was Sam Altman saying regular computer users shouldn’t have memory all memory should be allocated to data centers.
What will be the third crack in the dike?
Not selling them chips virtually guarantees that they will learn how to make their own, very quickly. The end result of that is much, much worse for the USA.
(and could leave Taiwan in a much weaker position in the near future)
For another example, the reason Chinese are not allowed on the 'international' ISS is because of the Wolf Amendment. [1] The idea was that if we just allowed them onto the ISS they'd steal all of our elite tech or something. Within a decade they had launched their own space station and carried out various revolutionary space activities.
And of course all of that happened while the ISS itself continues to deteriorate with no replacement in sight. And when China did an unprecedented sample return mission from the far side of the Moon and offered the whole world access to samples, US researchers had to jump through a bunch of hoops to receive them, again thanks to the same law.
It's just myopia.
Are you proposing we guarantee access to avoid potential access?
An economy kept dependent on American tech is unlikely to drop bombs on America and vice versa. Co dependency prevents war.
If the US or China developes an AGI that can leverage RSI, the game is over. It is over in the same way that WW3 is over as soon as the missiles are in the air: it doesn't matter who has the fastest or highest yield missiles, because there is a finite amount of time before there is nobody left to care about the advantages of one side. Except at least WW3 is over in about an hour, whereas an AGI could take years or decades to complete whatever thing it has been misaligned to desire.
AI alignment research is still in its alchemy era, and global cyber security is underfunded and bad. As a species, we simply do not have the tools to deal with AGI that can self-improve.
NVIDIA’s commercial interest is pretty clear… it’s beneficial to them. to sell everyone to maintain the dominance of CUDA. That is not necessarily the national interest. And remember that espionage and infiltration is not a one-sided affair, the US may well have the ability to sabotage Chinese development efforts. You just don’t know.
End of the day, corporate CEOs are very powerful, but they’re not as powerful as the government. My guess is that Jenson truly didn’t think that this individual was a spy, and figured that the questionable legality of chip exports to Chinese proxies could be addressed by ingratiating the administration.
No, they haven't. They spent the first decade after WW2 repairing the damage left behind by the Japanese and their Civil War. Subsequently they managed to starve tens of millions afterwards in several missteps, and then righted themselves.
They then spent three to four decades lifting about thrice the US population out of extreme abject poverty, and then in the last 20 years catching up. And catch up they have; they are already the leaders in green technology and EVs, already manufacture some of their own chips, already have indigenous military aerospace capability, which they are now expanding to civilian use. They make almost everything in the average household.
Where is the evidence that China is catching up considered a fiat d accompli?
Do you presume that China is destined to always at least catch up in every sector or field ever? That stinks of Chinese Exceptionalism to me.
If they could just "catch up" they would, and whether we sell them chips wouldn't matter because they would want to create their own Nvidia and export chips and make money too.
Suppose you want to build something that requires technical knowledge (a GPU, a fighter jet, ...). You have engineers, but they don't yet know how to build it.
If no one on earth has ever built this thing, you need to consider the possibility that it can't be done at a cost you can pay. Perhaps it violates some law of physics, or it can't be manufactured without knowing some chemistry/physics that we are 100+ years from discovering.
If a company in some other country is already building this thing, you can be sure it is possible to build it. That is a critical piece of information! It means you don't need to discover (much) new physics. It means you can buy/steal a working copy and look at it to learn about it. It means there are engineers you can hire/spy on to understand their design process. It means you can more easily convince a bank/government to fund development, as they have proof your goals are achievable.
Inventing a thing is much harder than inventing a clone of a thing. The idea that China doesn't have the people and organization to copy things other countries make is the opposite of Chinese Exceptionalism. It is the belief that the Chinese are uniquely incapable of engineering work. This is obviously not true.
The phrase is “fait accompli”, without a D, and is French for “accomplished feat”.
The ever closing gap between the capabilities of frontier models vs chinese open weight models.
Same when you sanction food imports. The elites who make decisions will always be fed. Sanctions are almost always an attack on the working class who do not make the high-level political decisions the sanctions are trying to address. Not to mention the idea of "making life hard on them so they revolt" almost never works.
Sanctions should be a bigger conversation. They are almost never needed and especially trade tariffs and such are always used for the most corrupt cronyism.
There’s also the minor quibble that restricting food exports to starve another country are illegal under international law.
For the record I think that less than half of the people who thought the dress was black and gold were being influenced by the Chinese government to take that position.
If the sale is not about money or share price then maybe it's ideological.