Notes - Oliver Hayman

Apr 2026

Models of Safety Evaluations of AI Deployment Protocols

12 Apr 2026 3 min read

Evaluating Language-Model Agents on Realistic Autonomous Tasks

10 Apr 2026 2 min read

Mar 2026

AI Control

24 Mar 2026 3 min read

DOPE Algorithm

14 Mar 2026 3 min read

Early work on monitorability evaluations

13 Mar 2026 1 min read

Dreamcoder

05 Mar 2026 8 min read

Provable Safe Reinforcement Learning with Binary Feedback

02 Mar 2026 4 min read

Feb 2026

Safe Exploration in Reinforcement Learning

17 Feb 2026 2 min read

Jan 2026

Rethinking Lipschitz Neural Networks

16 Jan 2026 1 min read

DoomArena Notes

14 Jan 2026 2 min read