OpenAI Capabilities Researcher Dan Selsam Publishes Personal Statement on AI Risk

austinc3301 · x · 2026-09-15

Dan Selsam, an OpenAI capabilities researcher since 2022, published a rare personal statement on AI risk, shared by former colleague @DKokotajlo. His background spans 15+ years: probabilistic programming languages at MIT, early development of the Lean Theorem Prover at Microsoft Research, pioneering neural networks learning to reason during his Stanford PhD, and chain-of-thought optimization work at OpenAI. He admits he "barely looks at raw code anymore" and struggles to maintain the discipline to engage deeply with the model's explanations and proposals throughout the day.

Related event: OpenAI researcher Selsam warns alignment evals are failing(20 posts)→

Original post →

More from AGI Musings

AGI Musings channel →