An ex-OpenAI researcher's influential nonprofit is now smack in the middle of AI's doom debate

An Anthropic researcher who wrote about "extinction-level risks" said he plans to join METR. This AI safety lab has become a prominent AI umpire.