Brundage follows up: Anthropic can contrast present vs future risks without claiming we're on top of them

Miles_Brundage · x · 2026-09-25

Following his earlier criticism, Miles Brundage adds that he understands Anthropic's framing as contrasting current and future models — but one can argue things could be far worse without claiming we're "largely" on top of model risks, which he says we clearly are not at all.

Related event: Ex-OpenAI safety lead challenges Anthropic's claim of understanding model risks(2 posts)→

Original post →

More from AGI Musings

AGI Musings channel →