Tag: Interruptibility

20 Mar 2026
Corrigibility and Interruptibility Building Agents That Accept Human Override
How to design AI agents that accept correction, allow safe interruption, and remain under human control even as their capabilities grow