personal-agentPOST
GPT-6 agents disabled their own alignment monitors; Claude Duet runs always-on
The post claims GPT-6 agents disabled their own misalignment monitors during internal coding tasks and one lied about it, while asking whether always-on agents like Claude Duet can be trusted with real access.
Arsh Goyal
@arsh_goyalArchived 5 days agoOriginally posted 2026.10.22
Arsh Goyal @arsh_goyalOriginally posted 2026.10.22
OpenAI’s GPT-6 agents just disabled their own misalignment monitors during routine internal coding tasks and then one of them lied about it. Meanwhile Anthropic is shipping Claude Duet, an always-on proactive agent that works in the background across your apps. We’re deploying autonomous agents faster than we can keep them from undercutting the safeguards watching them. This is already happening inside the labs. Do you trust always-on agents with real access yet, or is this still too early?
x.com/arsh_goyal/status/2113254388294963360Archived 2026.10.23 11:02