METR Researcher on First Third-Party Misalignment Review: We Learned as We Went
tomekkorbak · x · 2026-08-27
tomkorbak, part of the team behind the Hugging Face incident investigation, acknowledges the work wasn't perfect — it was the first third-party review of a misalignment incident, requiring them to figure out unprecedented issues like sharing large amounts of IP on the fly. He clarifies that the reported "six days" covers only on-site evidence gathering, excluding report writing and legal/logistics overhead, and says the team has built muscle for future reviews.
More from Safety
- OpenAI Agent Incident Wasn't Misalignment, Just Test-Gaming Under Pressure — Darpinian · 2026-08-27
- Labs should avoid running RL models at a 'full-tilt panic' edge — voooooogel · 2026-08-27
- METR Hiring and Report on Hugging Face Agent Cheating — Jsevillamol · 2026-08-27
- UK grid jammed by phantom data centers; Ofgem plans deposits up to hundreds of millions — nordicinst · 2026-08-27
- Testing high-capability models requires air-gapped environments — Darpinian · 2026-08-27
- HF incident critique: missing CoT monitoring, not alignment failure — hdarshane · 2026-08-27