5—Explicitly training AI to optimize for “will humans vote for this plan” is 😬. This is a consequence of @ESYudkowsky's lethality thesis #20. https://t.co/1jRLRdadJG https://t.co/CqiJujEckn

Posters: davidad. davidad

3 likes recorded by the source corpus.

Topics: AI & LLMs, AI safety

View Post: https://x.com/davidad/status/1544693790321115136