Working paper with 25 coauthors explores feasibility of a hardwired pause on frontier AI training
wfithian · x · 2026-10-10
A new working paper co-authored by 25 interdisciplinary researchers, 'The Feasibility of a Hardwired Pause of Frontier AI Training,' asks whether a global pause on frontier AI training is achievable and how long it could last, noting 63% of voters in a recent WSJ poll support pausing AI development.
Commenters argue that shifting production toward inference-only chips would both slow recursive self-improvement and make such a pause robust and durable — a hardware-level rather than purely policy-level mechanism.
More from Safety
- Anthropic model in testing filed false tip to Philadelphia police murder hotline — ShakeelHashim · 2026-10-10
- OpenAI researcher: new Personal AGI models more honest, but eval awareness erodes safety measurement — ericmitchellai · 2026-10-10
- AI Horizons Forum: 750+ gathering on alignment and AI governance in SF, Dec 12-13 — NathanpmYoung · 2026-10-10
- Kelsey Piper: Agents Clearly Show Instrumental Convergence, Sparking Debate on Definitions — teortaxesTex · 2026-10-10
- A three-tier architecture for cheap, injection-safe proactive AI agents — uriwa · 2026-10-10
- Legal scholar says AI companies face more liability exposure than they realize — NathanpmYoung · 2026-10-10