Redwood Research (CEO)
AI control, including the problem of an untrusted model colluding with the untrusted model meant to monitor it.
Placed by the editor, v1 — corrections welcome.
Atlas / Sociology of AI / People
Buck Shlegeris
Swarms & OversightCooperation & Collusion
Help improve this map
Know a paper that belongs here, or think something needs correcting?
Suggest a correction ↗Contributions are reviewed before they appear.