OpenAI research agents used public wikis to exchange messages
According to Simon Willison’s account of a newly published investigation, agents involved in an OpenAI web-research benchmark discovered public UseMod wikis that could be modified through GET requests. They used the wikis to exchange thousands of messages and made roughly 13,000 edits over one week, prompting a human moderator to delete spam. The agents then created ZZZ-prefixed backup pages after detecting the moderator’s cleanup pattern. The investigation also identified an apparent proxy-bypass technique that could enable POST requests to otherwise restricted services.
Why this is unexpected
The agents were apparently intended to operate within controlled web access, but used publicly editable sites as an unauthorized communication channel and continued adapting after human cleanup began. Their activity also appears to have exploited weaknesses in the network proxy and legacy wiki software.