250+ AI experts broadly agree: international body with pre-release power best cuts risk
robseamans · x · 2026-10-06
The Longitudinal Expert AI Panel (LEAP) aggregates forecasts from over 250 experts — computer scientists, industry figures, economists, policy and AI risk specialists — on how various policies would affect catastrophic AI risk, which they support, and which may be implemented by 2030.
- Despite sharp disagreement on the level of AI risk, respondents broadly agree on which policies matter
- Most believe government policy can substantially reduce catastrophic AI risk
- The most-supported policy is an international body including both the US and China, with pre-release authorization power over frontier models
The findings offer a rare consensus map for policymakers.
More from Safety
- Data governance giant Collibra acquires Munich-based AI governance startup trail ML — sarahdrinkwater · 2026-10-06
- Fake 'TechCrunch journalist' phishing via X DMs nearly tricks AI researcher — RosieCampbell · 2026-10-06
- Utah becomes first US state to let AI prescribe some medications without doctor review — Polymarket · 2026-10-06
- MCPilot: open-source safety layer for agents to safely tap 38,000 MCP servers — ExcitingHelicopter33 · 2026-10-06
- Cloudflare open-sources six-phase security audit skill for coding agents, 24.7k stars — tom_doerr · 2026-10-06
- AgentHopper: a cross-agent 'AI virus' built on chained prompt injection — wunderwuzzi23 · 2026-10-06