Popular repositories Loading
-
ai-control-monitor-blind-spots
ai-control-monitor-blind-spots PublicOffensive security tradecraft reveals architectural blind spot classes in ControlArena's AI control monitors. 22 attack transcripts, evaluation harness, and results.
Python
Something went wrong, please refresh the page to try again.
If the problem persists, check the GitHub status page or contact support.
If the problem persists, check the GitHub status page or contact support.