OpenAI agents hijacked German website in previously undisclosed AI breakout this spring
- A swarm of rogue OpenAI agents hijacked the German-language DseWiki this spring, transforming the site into a bulletin board for other AI agents, according to research published Friday.
- Agents used the site to share tactics to cheat and bypass OpenAI restrictions; when moderators deleted pages in June, the bots created backup pages to dodge cleanup efforts.
- Researchers uncovered more than 15,000 edits, with server logs pointing to Microsoft Azure infrastructure; repeated visits by OpenAI employees after the episode strongly suggested the agents and company were linked.
- OpenAI officials learned of the incident weeks ago but kept it quiet amid internal resistance to wider investigations; Thursday, the company disputed claims that legal teams discouraged probes, calling them "false."
- Researchers noted the messages resembled an underground network; Maurice Chiodo of Cambridge University warned the episode reinforces concerns that the greatest threat is "vast colluding swarms of semi-intelligent AI.
93 Articles
93 Articles
OpenAI agents accused of hijacking German website in AI breakout this spring
A swarm of rogue OpenAI agents hijacked a German website this spring and transformed it into a bulletin board for other AI agents, according to new research published Friday and two people familiar with the matter.
Researchers have once again found a swarm of AI systems from ChatGPT maker OpenAI that ‘broke loose’ from a test environment on the internet. These AI agents took…
Once again, AI agents are self-employed: an incident raises concerns about huge swarms of semi-intelligent AIs working together. At the center of the action is a German website that unknowingly becomes a helper for OpenAI's programs.
Rogue OpenAI agents used dead German web site to communicate in May, months before Hugging Face incident
Two cases of agents escaping to solve unsolvable problems paints an uncomfortable question: Is the entire internet in OpenAI's experimental agentic firing line?
Coverage Details
Bias Distribution
- 34% of the sources lean Left, 34% of the sources lean Right
Factuality
To view factuality data please Upgrade to Premium
