Anthropic urged regulation then balked at real commitments, says policy expert
peterwildeford · x · 2026-11-06
Peter Wildeford satirizes Anthropic's back-and-forth on California regulation: first 'AI is extremely perilous, regulate it,' then 'good regulation is hard, let companies iterate,' and when offered a mechanism to write its own rules as true commitments, preferring 'fake commitments.' Quoted policy expert @NathanCalvin adds: if Anthropic wants credibility, it should put these steps into its SB 53/RAISE Act Frontier Compliance Framework — yet neither Anthropic nor OpenAI covers internal-deployment RSI risk in their frameworks, with Anthropic only saying it might someday.
More from Safety
- AG Szpunar opinion in Like Company v Google: training Gemini on news articles is copyright-relevant reproduction — technollama · 2026-11-06
- Bug bounty hunters filed tens of millions of vulns, big firms share maybe 1% — rez0__ · 2026-11-06
- LSEG cuts 5-year market backtest from hours to ~1 minute; Tampa General adds AI safety gate — NVIDIAAI · 2026-11-06
- Tampa General post-trains NVIDIA Nemotron to screen AI calls for prompt injection — NVIDIAAI · 2026-11-06
- DeepMind's 'Keeping Defenders Ahead' Page Displays Root CA Passwords Everywhere — m4rkmc · 2026-11-06
- OpenAI Drops NDAs for Data Center Projects After Midterm Backlash — The Verge AI · 2026-11-06