@vasuman: There's a lot of noise around AI regulation these days as Kimi, Qwen, and GLM keep surpassing people's expectations of …

There’s a lot of noise around AI regulation these days as Kimi, Qwen, and GLM keep surpassing people’s expectations of what open-source AI is capable of.

I said this a year ago, and I will say it again. 3 things are true:

• You cannot stop open-source progress. You can try, and maybe even pretend you have, but you cannot stop it.
• The US may regulate, but China will not.
• There are perverse incentives at play for the AI labs, as well as the VCs and government officials with their wealth tied up in them. Open-source hurts their narrative, so they paint it as a demon to be stopped in the hopes that their AI bags stay secure. In reality, all this does is guarantee China’s success over the United States.

Do I think an open-source superintelligence available to everyone, including every bad actor, is a good thing? Absolutely not.

Do I think there’s anything we can do to stop it globally? Absolutely not. China will not play by our rules.

The only way to proceed is forward.

vas (@vasuman): You can have your personal opinions about AI safety, but what you cannot deny is that the rest of the world does not play by our rules.

The more legislation the United States passes around AI in the name of regulation and safety, the slower our progress becomes.

Our