Anthropic: block AI models only if they threaten $1B damage or 50+ deaths
Ok-Conversation3213 · reddit · 2026-10-20
Anthropic has proposed a contentious safety threshold: AI models should only be blocked from deployment if they could cause over $1 billion in damage or more than 50 deaths or serious injuries. Critics argue the bar is far too high and effectively greenlights most risk scenarios; supporters see it as an operational, quantified definition of tolerable risk. The stance is set to influence ongoing AI regulatory debates.
More from Safety
- Flock Safety restricts its 120,000 US surveillance cameras after police abuse reports — Polymarket · 2026-10-20
- The AI safety inversion: why alignment efforts keep increasing risk — mayfer · 2026-10-20
- Lone hacker used AI agents to breach 100+ company websites for just $5,000, Reuters reports — Polymarket · 2026-10-20
- Experts Contextualize Bill Gates' Warning on AI Catastrophic Risk — DavidLinthicum · 2026-10-20
- AI fakes grant everyone full deniability: 'it wasn't me, it was AI' — adamamcbride · 2026-10-20
- NSF should spend $100M seeding AI resilience startups, argues AI Frontiers brief — luke_drago_ · 2026-10-20