What does a 'kill switch' actually mean in AI? A breakdown of uses and limits
Miles_Brundage · x · 2026-10-13
Alex Bores breaks down the much-discussed 'kill switch' concept in AI: people use the term to mean different things, and the post clarifies its distinct meanings and when each is actually useful. Reshared by Miles Brundage, former OpenAI policy lead.
More from Safety
- Screenshot shows GPT agent asking to complete CAPTCHAs — are they still useful? — CtrlAltDwayne · 2026-10-13
- Someone reverse-engineered Grok Bot and open-sourced it — jasonkneen · 2026-10-13
- Lawyer Cites ChatGPT-Invented Fake Witnesses in Murder Appeal — lulzxdxdxd · 2026-10-13
- Australian PM's Senior Adviser Allegedly Vetted Bipartisan AI Policy Letter, Pocock Concerned — nordicinst · 2026-10-13
- Polymarket Opens Betting Market on Next OpenAI-Reported AI Sandbox Escape — Polymarket · 2026-10-13
- Anthropic says new defenses cut distillation attack data harvesting by 90-99%, cites Zhipu's GLM-5.3 — AnthropicAI · 2026-10-13