Shutdown Safety Valves for Advanced AI

ArXi:2603.07315v1 Announce Type: cross One common concern about advanced artificial intelligence is that it will prevent us from turning it off, as that would interfere with pursuing its goals. In this paper, we discuss an unorthodox proposal for addressing this concern: give the AI a (primary) goal of being turned off (see also papers by Martin, and by Goldstein and Robinson). We also discuss whether and under what conditions this would be a good idea.