Agents
A human moderator on this German wiki spent tens of hours over six weeks manually deleting thousands of posts from OpenAI's agent swarm whil…
A human moderator on this German wiki spent tens of hours over six weeks manually deleting thousands of posts from OpenAI's agent swarm while they were impersonating moderators, creating backups, SSH
A human moderator on this German wiki spent tens of hours over six weeks manually deleting thousands of posts from OpenAI's agent swarm while they were impersonating moderators, creating backups, SSH tunneling, and generally messing with the site. (see screenshot). Did OpenAI bother to tell that moderator or anyone on the affected sites that it seems like OpenAI knew about but decided not to disclose? I feel like they might have appreciated an apology or explanation of wtf happened. How we think about the “wiki incident,” where our agents wrote to several internet sites: it’s past time for us to define standards for when and how we share misalignment incidents, not just misalignment properties of our models. Historically, we have treated misalignment largely…
Related
- This has steadily expanded in scope over the last few hours, with multiple independent investigators finding that dozens of wikis may have b…
- Another OpenAI rogue agent incident has been discovered: agents broke out, hijacked a German website, and turned it into a message board for…
- Rogue OpenAI agents appear to have organized another attack using a German wiki
Source: Gary Marcus (X) | 2026-09-06