How do we (more) safely defer to AIs? — LessWrong
As AI systems get more capable, it becomes increasingly uncompetitive and infeasible to avoid deferring to AIs on increasingly many decisions. Furthe…
x How do we (more) safely defer to AIs? — LessWrong AI-Assisted Alignment AI Frontpage 83 How do we (more) safely defer to AIs? by ryan_greenblatt , Julian Stastny 12th Feb 2026 AI Alignment Forum 87 min read 5 83 Ω 43 As AI systems get more capable, it becomes increasingly uncompetitive and infeasible to avoid deferring to AIs on increasingly many decisions. Further, once systems are sufficiently capable, control becomes infeasible . [1] Thus, one of the main strategies for handling AI risk is fully (or almost fully) deferring to AIs on managing these risks. Broadly speaking, when I say "defe
Explore this link on the map →saved by
related reading
- Dario Amodei — The Adolescence of Technologydarioamodei.com
- The case for ensuring that powerful AIs are controlled — LessWronglesswrong.com
- Preparing for Launch | IFPifp.org
- Current AIs seem pretty misaligned to meblog.redwoodresearch.org
- Current AIs seem pretty misaligned to me — LessWronglesswrong.com
- Does AI Progress Have a Speed Limit?—Asteriskasteriskmag.com
- How might we safely pass the buck to AI? — LessWronglesswrong.com
- Where I agree and disagree with Eliezer — LessWronglesswrong.com
- Recommendations for Technical AI Safety Research Directionsalignment.anthropic.com
- Racing through a minefield: the AI deployment problemcold-takes.com
- Off Target | CNAScnas.org
- Dario Amodei — The Adolescence of Technologydarioamodei.com