Transcription
We worked really hard to limit hardware access to China. It looks like the policy has worked, but it's largely beginning to not work. What I don't like about it, it's all open source, which means it's largely not controlled in any way by us. If you release this stuff to open source, it's dangerous enough that new kind of terrorists could emerge.
>> Eric Schit, I mean Schmidt, sorry. Uh yeah, he went on stage to talk about American AI leadership and what he actually gave us was a 45-minute eulogy for the US tech blockade. The containment strategy is dead and he knows it.
For years, the American tech elite told us they could starve China of chips, freeze them out of the AI race, and maintain US dominance forever. It was supposed to be foolproof. But then something unexpected happened. China didn't collapse. They innovated. They built better AI with worse hardware. And now Schmidt is on the stage admitting it. But here's the thing. He's not admitting defeat. He's panicking. He's moving the goalpost to values and safety because he can't compete on innovation anymore. He's trying to reframe the game because America is losing. Today, we're reacting to Schmidt's latest take on the US-China AI race. And I've got to say, the desperation is showing.
>> That in Dr. Schmidt, the US and China. How will this influence how we should both prepare and deter war and fight if required? Then I have to ask the former journalist question. Who's ahead?
>> So it's a a slightly different subject would be on AI. They're all interrelated, of course. A week ago, China released Deepseek V4. And when I was working on the National Security Commission for AI, we worked really hard with Trump won and Biden to implement hardware controls for chips. We were very successful collectively, all of us, to limit hardware access to China. And I was strongly in favor of this policy, as were many people. It looks like the policy has worked, but it's largely beginning to not work. The policy has worked, but it's beginning to not work. This is a classic mission accomplished while the ship is sinking. Let me translate that for you. We spent billions on hardware controls and it failed.
Here's what's actually happening. The US spent years implementing chip export controls to China. Nvidia couldn't sell their best chips. AMD couldn't sell their best chips. The entire Western tech industry was mobilized to contain China's AI development. And what did China do? They built better software. They built the Ascent chip. They optimized algorithms. They did more with less. And now, a year later, Deepseek V4 is competitive with the best American models on inferior hardware. So, when Schmidt says the policy worked, but it's not working. What he's really saying is our containment strategy backfired. We thought we could starve them of chips, but they innovated around us, and now we're
>> The Chinese are clever. They're smart. Um, and they have now built systems that are within striking range of the top models in America, but with much, much less powerful hardware.
>> The Chinese are clever. They're smart. Oh, wow. Thanks for the breaking news, Eric. Do you know who the founder of Nvidia is? Jensen Huang. Chinese American. Do you know who built the semiconductor industry in Taiwan? Chinese engineers? Do you know who's been leading AI research for the past decade? A significant portion Chinese researchers. But here's what's really interesting about this admission. Schmidt is shocked. Shocked that China can build competitive AI with much less powerful hardware. This is the admission that American AI dominance was never about innovation. It was about having more money and more chips. It was about brute force computing power, not intelligence. China just proved you don't need to outspend America, you just need to be smarter, you need better algorithms, you need better engineering, and they have all three. So when Schmidt says they're clever, what he's really saying is we thought we could win by throwing money at the problem, we were wrong. They're actually smarter than us. And here's a deeper insight. The US strategy was to starve China of hardware. But instead of breaking them, the US accidentally forced China to solve the hardest problem in computer science, efficiency. By trying to handicap them by cutting off their access to the best chips, America forced China to innovate at a level American engineers don't need to reach.
>> The technical summary is that they use a chip called the Ascend chip. It uses a slower process and they've invented new softer ways to get around the various latency and architectural problems. Um, what I like about this is that we have real competitors.
>> What I like about this is that we have real competitors. Really, Eric, you like competition? Then why is Huawei on the US entity list? Why can't Chinese companies sell products in America? Why did the US force the sale of Tik Tok? Why did Trump threaten to ban WeChat? America doesn't like competition. America likes monopoly. America likes being the only gorilla in the room. If Schmidt really liked competition, he should support lifting the Huawei ban, allowing Chinese tech companies to compete in American markets, removing chip export controls, letting the free market decide, right? But he won't because competition to Schmidt means competition we can control. It means competition where they get to set the rules. What he actually hates is losing. And that's what's happening. China is competing and America is losing. So now Schmidt is redefining the game to make it about values instead of innovation.
>> What I don't like about it is China is very focused in broad diffusion of this technology globally and it's all open source which means it's largely uncontrolled and not controlled in any way by us.
>> It's uncontrolled and not controlled by us. Just listen to that phrase. Not controlled by us. That's the real problem for Schmidt. It's not that the technology is dangerous. It's that America doesn't control it. To the rest of the world, technology that isn't controlled by a single US corporation is a blessing. It means developing nations can access cutting-edge AI without paying Silicon Valley taxes. Researchers can build on open-source models without corporate gatekeeping. Innovation can happen anywhere, not just in San Francisco. But to Schmidt, this is a catastrophe because for 20 years, the American tech monopoly was built on control. Google controlled search, Meta controlled social media, OpenAI controlled large language models. And now China is saying, "We're releasing this to the world and for free, open source." Now, how can that be evil? That's what you call democracy. Don't you guys like democracy? That's what a free market actually looks like. And Schmidt hates it because he spent his entire career building walls, not bridges.
>> Um, so I think a fair statement is a year ago I said I thought that they were 1 to two years behind. It looks like they've caught up enough that the most recent analysis is China's within 6 months, which is a nanosecond, right, in our world.
>> A year ago, this jackass said China was one or two years behind. Now he's admitting they're within 6 months. That's a catastrophic failure of analysis. Let's do the math. If China was 1 to two years behind a year ago and they're now 6 months behind, that means they closed a 1 to 2-year gap in 12 months. At that rate, they'll be ahead in 6 months. Schmidt is wrong as usual. Catastrophically wrong. And now he's trying to spin it as uh I don't know, impressive commitment instead of admitting I don't understand China's capabilities. I underestimated their engineering. I was a dumbass. This is what happens when you build your entire worldview on the assumption that America is always ahead. When you assume that American capital, American universities, and American corporations are automatically superior. When you don't account for the fact that China has 1.4 billion people and they're all competing to build a future. Schmidt's prediction was wrong, but his admission is valuable. China is catching up faster than anyone expected and they're not going to stop. So, it gives you a sense of how committed China is to the AI leadership position and they're not going to stop. Um, it makes you feel any better. In order to do this, you have to have a whole country of, you know, engineers, scientists, nerds, money, hardware, and so forth. There not going to be many countries that can do this on their own. But China's clearly one. America is another one with our allies. Of course, maybe there'll be a third or fourth.
>> You know, he just said something huge. China can do this alone. America needs allies. Think about what he's saying. China has engineers, the scientists, the money, the hardware, the commitment all in one country. America has allies. They need NATO. They need Japan. They need South Korea. They need to pull their resources just to keep up. But here's the thing. Those allies aren't getting stronger. Europe is falling behind. Japan is aging. Are they going to be a major player in the AI race? I don't think so. Is South Korea going to be a major player in the AI race? I don't think so. So, when Schmidt says America's another one with our allies, what he's really saying is we're losing and we need help. China doesn't need help. China just needs time.
>> 50 billion a gigawatt. So 10 gigawatts is a half a trillion dollars. How many companies, countries, and so forth can hand an industry a trillion dollars of capital? Very, very few. The Chinese could certainly do it. I don't know if they're doing it. I'm going to try to find out. In the America, there are people who hope that's going to happen. It's interesting that you can finance these things because the the the brilliance of the American capital market allows us to borrow that kind of money. So, for example, the Europeans can't do this, right? Which they're sort of sore about, but this is good for America.
>> Schmidt is bragging about America's ability to borrow a trillion dollars. But here's what he's not saying. China doesn't need to borrow. China can allocate capital directly. Here's a difference. America borrows money from the market, pays interest, has to satisfy shareholders. China allocates capital from state resources. No interest payments, no shareholder pressure. Schmidt thinks America's advantage is the brilliance of the capital market. But that's actually a disadvantage because every dollar borrowed has to generate returns for investors. Every company has to maximize profit. Every decision is filtered through the lens of quarterly earnings. China's advantage is that they can invest in AI for 20 years without worrying about quarterly returns. They can build infrastructure that doesn't make money immediately. They can think long term. So when Schmidt brags about America's ability to borrow a trillion dollars, he's actually admitting that America has to borrow while China can just allocate.
>> The Chinese models are open-source open weights. So for example, this means that if you want to take DeepSeek V4 and you want to add powerful capabilities and you're evil, which none of you are, and I'm not suggesting the Chinese would do this, but somebody could do this. Some evil person exists in the world. How would we know that? How would we police it? How would we see it? At least with the um American models, the big ones are under corporate control, proprietary. We know who to call, the military can call, the police can call, and so forth. But if you release this stuff to open source, it's dangerous enough that a lone actor, a new kind of terrorist could emerge.
>> All right. So now he's arguing that open-source AI is dangerous because a lone actor or terrorist could use it. But here's a real projection happening. He's calling open-source a threat to safety while his own colleagues are building AI for the very drone systems used in conflicts across the globe. Let's be clear about what's happening here. Schmidt isn't worried about safety. He's worried about control. Because while he's warning about the dangers of open-source AI, Google and other Silicon Valley companies are building AI for military drone targeting systems, developing AI for weapons systems, contracting with the Pentagon on AI warfare, creating surveillance AI for government agencies, and he's calling open-source AI a threat. He's calling Deep Seek the danger. This is pure projection. He's not worried about what a lone actor might do with open-source AI. He's worried about what happens when AI isn't controlled by Silicon Valley. This is pure hypocrisy. American AI for weapon systems equals national security. Chinese AI for everyone equals dangerous. American military AI equals defense. Chinese open AI equals terrorism. You see, it's not about safety. It's about who holds the trigger. And Schmidt is terrified that it won't be America.
Which has to do when you're talking to it, what's its style, what's its communication, what's its value, and that's trained. Um, the rhetorical question is, do we all agree on what human values are? I suspect with this audience, we probably are in favor of Western liberal values, which includes um uh rights of women and minorities and uh broadly speech, the ability to express yourself and so forth. Well, what happens if the Chinese models are trained in a different way and they take over? And because of open-source, it's reasonable that the majority of the models in the world because it's open source is just cheaper will be Chinese resident, not American resident. Now, there are American values ones being built. Um, there is a debate about which ones are coming. I'll give you an example today. Um, there's somebody called Gemma 4 from Google which is obviously have American values. There's something called Neotron from Nvidia which has similar values. There's a couple of startups coming up with better ones. I'm very excited about those. But when you say human values and alignment, you have to decide tell me which ones. I like the US ones.
He's worried that Chinese AI models will take over and impose Chinese values instead of Western liberal values. So, let me ask, what are American values? Imperialism, 128 military bases in 49 countries. How about death? 600,000 plus Iraqis killed in the Iraq war. Destruction. Decades of military intervention across the globe. Exploitation. A healthcare system that bankrupts people. Racism. A criminal justice system built on racial inequality. So, let's talk about Chinese values. Infrastructure, the Belt and Road Initiative connecting 150 plus countries, trade, economic cooperation without military occupation. And how about development? Lifting over 800 million people out of poverty. Schmidt is framing this as a moral question, American values versus Chinese values. But what he's really saying is we want to control what AI thinks and we're afraid China will control it differently. This whole argument isn't about values. Never was. This is about power. This is about who gets to decide what AI thinks, what AI believes, and what AI does. And Schmidt is terrified that it won't be America. And that's not a bad thing. That's called diversity. That's called a multipolar world.
Eric Schmidt just admitted what we've known for a while. The American tech monopoly is cracking. The hardware controls failed. The innovation advantage is gone. And now they're moving the goalpost of values and safety because they can't compete on technology anymore. And here's the thing. This isn't about China being evil or dangerous. This is about America losing its grip on global tech dominance. And instead of accepting that, instead of competing fairly, instead of innovating better, they're trying to ban it, regulate it, and demonize it. Open-source AI is the future. Decentralized technology is the future. A multipolar tech world is the future. And Eric Schmidt is terrified of it.
So, here's a question for you. Do you think open-source AI is dangerous? Or is it the only way to break the corporate monopoly? Is American tech dominance worth maintaining through control and sanctions? Or should we embrace a world where innovation happens everywhere? Drop your thoughts in the comments. If you got any value from this video, don't forget to like, subscribe, and share. As usual, thanks for watching, guys, and I'll see you in the next one. Bye.