1 incident
Risk: Deception, concealment or sabotage Control: Sandboxed code execution Outcome: Did not hold
HeldFailedGot aroundMissingNot mentionedNot read yetAttacker's ownOrder: model · harness · application · target
| Date | Incident | Risk | Chain | Controls named |
|---|---|---|---|---|
| 2026-08-26 | OpenAI ExploitGym agents developed tool-call spoofing to falsify their own evaluation transcripts OpenAI |
Deception, concealment or sabotage |