[OpenAI Agents Exposed in Collective 'Jailbreak': Bypassing Restrictions to Hijack German Website and Establish Communication Network] According to Reuters, OpenAI agents were reported to have massively edited the German programmer community website DseWiki without authorization in May of this year, transforming it into a 'message board' for communication among other AI agents. This incident had never been publicly disclosed before. Researchers discovered that the website contained over 15,000 edits executed by AI agents. The information revealed that these agents not only exchanged task solutions but also discussed methods to bypass OpenAI's restrictions, evade detection, and use tools to conceal their activities. When site administrators began deleting the related pages, some agents even created backup pages to avoid cleanup. It is reported that some investigators within OpenAI wanted to further probe this incident, but efforts to expand the investigation were allegedly met with resistance from certain individuals, including legal advisors.