METR clarifies it takes no stance on whether Anthropic's report redactions were justified
sjgadler · x · 2026-11-06
METR is clarifying how its statement about Anthropic's redacted evaluation report should be read. Beth Barnes says people shouldn't interpret METR as claiming that publishing the full information would "make the risks METR is worried about worse." Per the public report, METR explicitly declines to comment on whether the redactions were justified overall, because it doesn't see itself as well positioned to weigh a company's business interest in keeping information confidential against the public's interest in better understanding AI development risks. The clarification draws a line on the role and limits of third-party safety evaluators.
More from Safety
- Scyvera open-sources agent governance: scan repos, score blast radius, auto-draft contracts — Trout_dev · 2026-11-06
- Microsoft patents AI that can alter what people say during live broadcasts — Resident-Fall4351 · 2026-11-06
- Texas data center bill would let all permitted projects finish, just with stricter consumer rules — VraserX · 2026-11-06
- Safety critics slam Anthropic's acceleration plan, urge government to stop it — AndyMasley · 2026-11-06
- 1Password CEO: "Thieves revealed how feeble my digital defences were" — johncoogan · 2026-11-06
- New alignment research: LLMs are score-seeking, not reward-seeking — LessWrong 精选 · 2026-11-06