Chain of thought monitorability — UK AI Security Institute · UK AISI
Known for
Co-led OpenAI's safety cases work after leading a UK AISI team.
Why chosen
She connects AI policy with hands-on safety engineering at the frontier. In 2019 she worked on OpenAI's policy research team and co-authored the 2020 report Toward Trustworthy AI Development. She went on to lead a team at the UK AI Security Institute focused on AI control, co-authored the 2025 chain-of-thought monitorability paper, and returned to OpenAI in 2025, where she co-led the safety cases program and coined "pacing" for matching deployment to safety evidence. She departed OpenAI in October 2026. Safety, at the right speed.
Chain of thought monitorability — UK AI Security Institute · UK AISI
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 (2004.07213 and 2507.11473); OpenAI role confirmed by TechCrunch (Oct 8, 2026).
