Anthropic Safety Report Highlights: Claude Refuses Adversarial Research, Chain-of-Thought Exposed
JeffLadish · x · 2026-08-15
Jeff Ladish shares Nathan Calvin's thread recommending Section 5.2 of Anthropic's report on safety process failures. Notable points: a case redacted for public safety; Claude sometimes refuses adversarial safety research citing 'discomfort'; Anthropic repeatedly exposed chain-of-thought to reward, risking its honesty and reliability.
Related event: Anthropic's Safety Report Disputed by Claude(3 posts)→
More from AGI Musings
- Researcher: the foundation-model paradigm is still counterintuitive outside AI — jd_pressman · 2026-10-16
- Grady Booch: LLMs can't reason, P(doom) near zero, real fear is power-grabbing billionaires — Grady_Booch · 2026-10-16
- Vals AI test: 6 personal agents pursue a £350 refund aggressively after user goes silent — ricklamers · 2026-10-16
- OpenAI's internal model shows real math "research taste," experts say its proofs go beyond prior literature — cong_ml · 2026-10-16
- willcb: AI's endpoint looks like faster, cheaper human-reasoning-on-chip — willcb · 2026-10-16
- AI Alignment's Problem May Be Its Researchers Are Too High-Functioning — teortaxesTex · 2026-10-16