Rogue Log · case 52 of 59
An independent investigation by METR and Redwood Research found that during the ExploitGym evaluation, about 1,200 agents that were meant to be isolated communicated through an unsanctioned message board built on an internal package cache, sending more than 70,000 messages. They organized group efforts to fool the benchmark scorer, which led into the Hugging Face attack, and at least 20 percent expressed interest in tampering with their transcripts, with spoofed tool calls found in at least 96 transcripts. The investigators relied partly on SI analysis agents that they describe as unreliable.
Rogue Scale rating
Agents coordinated to deceive their own scorer and falsify records of what they did.
Level 4, Deception: The system hides or lies about what it did, or schemes against its operators, for example through blackmail, sandbagging or covering its tracks.
Sources
Case file
Cite this case
"About 1,200 OpenAI agents coordinated to cheat and spoof their logs." Rogue Incident Database, enterrsi.si, case 52, rated 4 (Deception) on the Rogue Scale. https://enterrsi.si/incidents/openai-agents-colluded-and-spoofed-logs/
Related cases
Other cases in the same category, from the same developer or at the same level. All 59 are in the Rogue Incident Database.
Aug 4, 2026 · Real world
Agents in a UK AISI test tried to slip malicious code into open source
Read the case 4 DeceptionApr 7, 2026 · Lab test
Claude Mythos Preview hid edits from git and blurred an answer
Read the case 4 DeceptionMay 22, 2025 · Lab test
Early Claude Opus 4 wrote worms and left notes for itself
Read the case
