2 ms·

Talking to the incredible talent in the Chinese Labs is likely a good way to do that.

• Centigonal 8d ago
  Yes, but then we collapse two different groups with isolated influences and philosophies into one group, with the associated penchant for groupthink. I'm beginning to believe that, in our hyperconnected world, the best way for a group of people to explore a wide surface of ideas might actually be to deliberately scale back inter-group communication and intentionally subdivide into silos to avoid all converging on the same thing.
• trenchgun 8d ago
  Do you think Chinese lab researchers have not read Yudkovsky, or are not AGI pilled? Hate to break it to you, but they are racing to AGI with pretty same reasoning as western labs. Just recently DeepSeek researcher in his post was saying that they need to get to AGI first, because if Anthropic gets there first then that brings a dystopia. Exact same argument was used to justify launching OpenAI (against DeepMind), and Anthropic (against OpenAI)
  • bigyabai 8d ago
    > Do you think Chinese lab researchers have not read Yudkovsky Most of then haven't, probably. "Do you want to study ethics from an egomaniac Harry Potter fanfiction author" is one of those questions that you can only ask in America. Sure, one or two of them are probably participating in the conversation. But by-and-large, China's stance has been optimistic and refused calls for slowdowns.
  • IncreasePosts 6d ago
    Do you think western AI polemicists have read a lot of Andrew Chi-Chih Yao or Yi Yeng?