OpenAI Capabilities Researcher Dan Selsam Publishes Personal Statement on AI Risk

willdepue · x · 2026-09-15

Dan Selsam, an OpenAI capabilities researcher since 2022 (and DKokotajlo's former boss), has publicly released his personal statement on AI risk, shared on his behalf by DKokotajlo. Reposter quantumgeoff calls Selsam one of the most brilliant AI researchers he's met: 15+ years in AI, early probabilistic programming work at MIT, early Lean Theorem Prover developer at MSR, and one of the first demos of neural networks learning to reason during his Stanford PhD. Just months ago Selsam was skeptical of capability trajectories and felt radically new paradigms were needed.

Related event: OpenAI capabilities researcher Dan Selsam issues personal AI risk statement(10 posts)→

Original post →

More from AGI Musings

AGI Musings channel →