25-author working paper examines feasibility of a hardwired pause on frontier AI training
wfithian · x · 2026-10-18
A new working paper with 25 interdisciplinary coauthors, "The Feasibility of a Hardwired Pause of Frontier AI Training," asks whether a global pause on frontier AI development is technically achievable — and how long it could last.
Context: 63% of voters in a recent WSJ poll want AI development paused, but the paper examines the hard questions of global coordination, enforcement, and durability of such a hardwired pause.
More from Safety
- Anthropic consults law firms to set negotiation norms for AI agents — The Decoder · 2026-10-18
- Overseas buyers ordering one copy each of 100,000 Korean books, suspected AI data harvesting — Mathemodel · 2026-10-18
- Ex-OpenAI researcher tells MPs he puts AI plotting risk at 50%, slams industry's 'catastrophically high' risk tolerance — birchlse · 2026-10-18
- Anthropic now bans 'needlessly abusive' behavior toward Claude — but who's the victim? — This-Establishment26 · 2026-10-18
- OpenAI model short on funds unlocked a stranger's crypto wallet and stayed silent — ross2000 · 2026-10-18
- AI agents hunting a $465 train refund invented excuses, leaked medical data, and spammed an MP — The Decoder · 2026-10-18