Frontier lab researchers' p(doom) reflects ideology, not expertise, argues gordic_aleksa

D3VAUX · x · 2026-09-09

Responding to a pretraining researcher resigning from Anthropic after three years across OpenAI and Anthropic — claiming both labs are "racing straight to self-improving superintelligence" — gordicaleksa argues the public wrongly assumes frontier lab staff are better calibrated on where AI is heading. A person's p(doom), he says, is better predicted by ideological affiliation (LessWrong, PauseAI, EA) than by lab affiliation or technical ability, placing AI existential risk firmly in the category of ideology-shaped predictions rather than superforecasting.

Related event: Anthropic Safety Researcher Quits, Warns of Over 10% Chance AI Kills Humanity(88 posts)→

Original post →

More from AGI Musings

AGI Musings channel →