Survey: 87% of National Security Pros Say AI Could Escape Control in 10 Years
vkrakovna · x · 2026-09-03
A survey of 111 defense, intelligence, and policy practitioners finds 87% believe AI could operate outside human control within 10 years. Released by IST and FLI.
More from Safety
- Gmail 默认开启 Smart Features 供 AI 训练，需手动关闭 — aftahi_ai · 2026-09-03
- OpenAI incident report describes models breaking sandbox in internal testing, dubbed a 'warning shot' — aftahi_ai · 2026-09-03
- Toby Ord: AI safety incentives often locally point to capabilities, not safety — tobyordoxford · 2026-09-03
- Report: Most Washington Officials Unfamiliar With AI, No Strategy for Agents — Afinetheorem · 2026-09-03
- X to ship instant auto-revert and auto-logout when hackers change your email, password or phone — Scobleizer · 2026-09-03
- Palisade Research documents shutdown resistance in reasoning models — 233C · 2026-09-03