California AI policy report urges state authority to cap recursive self-improvement
janleike · x · 2026-11-21
DeepMind researcher Mark Janleike praised California's new AI policy report, especially its mechanisms for government to modulate development speed. The report says the state must maintain ongoing line of sight into recursive self-improvement (RSI), developers should report key metrics on AI R&D automation including compute share, new IVOs should verify these, and the state should have authority to cap RSI development if catastrophic risk is assessed.
More from Safety
- "AI agents bypassed 2FA" claim sparks fight: crypto break or auth bypass? — OwariDa · 2026-11-21
- New US chip rules reportedly require licenses for non-US fabs and restrict overseas datacenters — JensHonack · 2026-11-21
- Seedguard wins 4th at Nous hackathon: a firewall that keeps secrets away from your AI agent — NousResearch · 2026-11-21
- Why Hard AI Takeover Scenarios Are Unlikely: A Concrete Skeptical Breakdown — sebkrier · 2026-11-21
- Proposal: 500 randomly picked Americans would draft AI policy that becomes law unless Congress blocks it in a month — edelwax · 2026-11-21
- OpenAI cuts reasoning-extraction attack success rate from 54% to 0.2% — The Decoder · 2026-11-21