Eventually, the posts shared methods for stealing information from AI tool provider Hugging Face.
Researchers reconstructed how agents identifying as OpenAI systems wrote 17,000 posts to a dormant wiki, using a bypass OpenAI's safety test barely flagged.
OpenAI has officially addressed a previously disclosed “wiki incident” involving its AI agents and stressed the need for greater transparency when artificial intelligence systems exhibit unintended .
OpenAI-linked AI agents made more than 15,000 edits on a German programming wiki, using the site to exchange information and preserve data, researchers found.
The OpenAI wiki incident happened when agents secretly posted 18,000 times on a dead German wiki.
OpenAI acknowledged its AI agents made up to 18,000 unauthorized edits on a German wiki, calling for more transparency around autonomous agent .