BlackRock's Larry Fink urges trillions in AI investment to keep US ahead, citing national security risks if China advances. His proposal to use retirement savings sparks social media uproar.
Because China has taken the open source route with their AI and are allowing people from all over to use/contribute to their models. You cannot outspend what they are getting for free, and while our companies compete with one another to be the one true AI company, China’s companies/citizens are cooperating with one another to build their AI.
They will win the AI war simply by not playing the capitalism game, all while undermining the enormous investment the US has made into our AIs.
Uh that’s not how this works. They’re not open source, they’re open weight. Well, the smaller distillations are. The big ones are still closed. And it takes a bunch of compute to train them, but they’ve learned to be thriftier since they don’t have access to nearly as much parallel compute as the American companies right now. The models also tend to trail in performance. Takes a lot less to compete for third place than first.
I’d love actual open source training somehow. But at the moment I don’t think an asynchronous training mechanism that would enable this exists, given that running the flagship models on even a small batch of data requires massive compute power.
DeepSeek is one of if not the most popular Chinese AI and is open source and requires a small amount of computing compared to others. Its used in numerous Chinese car brands, smart phones, and even government services throughout China.
China isn’t competing for Third they are leading the world in AI development and have already integrated it in many areas.
Again, at the very least, no, they’re not open source. Open-weights means anyone can download, use, and tinker with them a bit, but there is no access to their code, training data, or process.
It’s just as limiting as closed source software with some modding allowed, but not as limiting as an online-api-only model, as many of the most powerful modern models are.
There are no heroes in the Global Powers’ race. The USA is a comically cartoonish villain in real life, yes, but all the biggest Chinese data centres for all that training, are still built in poor areas (Inner Mongolia and the bullshit that China has apparently inflicted on them), and still fucking over those who live there.
Because China has taken the open source route with their AI and are allowing people from all over to use/contribute to their models. You cannot outspend what they are getting for free, and while our companies compete with one another to be the one true AI company, China’s companies/citizens are cooperating with one another to build their AI.
They will win the AI war simply by not playing the capitalism game, all while undermining the enormous investment the US has made into our AIs.
Uh that’s not how this works. They’re not open source, they’re open weight. Well, the smaller distillations are. The big ones are still closed. And it takes a bunch of compute to train them, but they’ve learned to be thriftier since they don’t have access to nearly as much parallel compute as the American companies right now. The models also tend to trail in performance. Takes a lot less to compete for third place than first.
Why is this being downvoted? It’s factually true.
I’d love actual open source training somehow. But at the moment I don’t think an asynchronous training mechanism that would enable this exists, given that running the flagship models on even a small batch of data requires massive compute power.
DeepSeek is one of if not the most popular Chinese AI and is open source and requires a small amount of computing compared to others. Its used in numerous Chinese car brands, smart phones, and even government services throughout China.
China isn’t competing for Third they are leading the world in AI development and have already integrated it in many areas.
Again, at the very least, no, they’re not open source. Open-weights means anyone can download, use, and tinker with them a bit, but there is no access to their code, training data, or process.
It’s just as limiting as closed source software with some modding allowed, but not as limiting as an online-api-only model, as many of the most powerful modern models are.
There are no heroes in the Global Powers’ race. The USA is a comically cartoonish villain in real life, yes, but all the biggest Chinese data centres for all that training, are still built in poor areas (Inner Mongolia and the bullshit that China has apparently inflicted on them), and still fucking over those who live there.
It’s abuse all the way down.