Tomek Korbak - Chain of Thought Monitorability for AI Safety [Alignment Workshop] · FAR.AI / YouTube
Known for
Co-first author of the chain-of-thought monitorability paper.
Why chosen
He connects cognitive science and philosophy with keeping AI agents honest. After a PhD at the University of Sussex on learning from human feedback, including time as a visiting researcher at NYU, he presented an oral at ICML 2023 on pretraining language models with human preferences. He worked on honesty training at Anthropic and on AI control and safety cases at the UK AI Security Institute, and in 2025 was co-first author of the cross-lab paper on chain-of-thought monitorability. At OpenAI he worked on monitoring agents for misalignment until October 2026. Reading the reasoning, so we can trust it.
Tomek Korbak - Chain of Thought Monitorability for AI Safety [Alignment Workshop] · FAR.AI / YouTube
Role
AI Safety Researcher, Formerly OpenAI
Signal
LeftDeparted OpenAI in October 2026; co-wrote an open letter calling for clear rules on outside safety collaboration. Authorship confirmed on arXiv (2507.11473); OpenAI role confirmed by TechCrunch (Oct 8, 2026).