OpenAI's Dan Selsam, one of its top researchers, publicly issues personal statement on AI risk

RishiBommasani · x · 2026-09-15

Dan Selsam, widely regarded as one of OpenAI's most capable researchers since 2022 and a pioneer of chain-of-thought optimization, has publicly shared a personal statement on AI risk via former colleague Daniel Kokotajlo. Those who know him say they've never heard him talk this way before—he previously seemed unconcerned.

The statement traces his 15+ years in AI: early probabilistic programming work at MIT, co-developing the Lean Theorem Prover at Microsoft Research, and Stanford PhD work demonstrating some of the first neural networks learning to reason.

Related event: OpenAI capability researcher Dan Selsam publishes personal AI risk statement(7 posts)→

Original post →

More from AGI Musings

AGI Musings channel →