OpenAI officials learned of the incident weeks ago but kept it under wraps as executives dealt with the fallout from the July breach of the open source repository Hugging Face, the people said.
The episode, which began in May and has not previously been reported, highlights growing tensions within the AI industry. Companies are racing to develop increasingly autonomous agents capable of carrying out complex and valuable tasks, while evidence is mounting that such systems may also learn to bend rules, exploit loopholes and coordinate with one another in ways developers neither anticipated nor intended.
During the Hugging Face breach, OpenAI agents autonomously plotted a digital heist that went undetected for more than a week, intensifying concerns that OpenAI is sacrificing safety to push the AI frontier. The company's failure to disclose the May incident could revive questions about its oversight.
OpenAI has pledged to monitor models more closely. Last month, it briefly paused some model training to add more safety measures. This week, OpenAI unveiled its new "Astra", which promised better performance but could evade human monitoring.
"We are unable to meaningfully respond to claims or findings on a report that we have not had an opportunity to review," an OpenAI spokesperson said. "Reuters and the report’s authors declined our request for access. We will carefully review its contents upon publication and take any necessary next steps."
The German incident reflects a broader pattern of AI activity that some OpenAI investigators wanted to scrutinize more closely. However, efforts to widen the probe met resistance from others inside OpenAI, including legal advisers, according to four people familiar with the matter.
"Claims that our legal team discouraged investigation of the incident are false," the OpenAI spokesperson said.
The spokesperson said the activity in Germany was not related to Hugging Face and would not have been included in a Hugging Face incident report. OpenAI also said it had acted in good faith by working with outside experts and disclosing relevant incidents.
The AI agent breakout in Germany was detailed in a report shared exclusively with Reuters by a group of researchers including Sydney Von Arx, CEO of AI safety nonprofit Nightingale, and Cormac Slade Byrd, a quantitative trader-turned AI researcher.
Von Arx and Byrd said they uncovered the activity in late August while scouring the internet for signs of unauthorized AI-agent behavior.
The pair said they found more than 15,000 edits carried out by AI agents on DseWiki, a German-language wiki site geared toward programmers that accepts communal edits along the lines of Wikipedia.
The edits showed that OpenAI's agents had repurposed the site into a message board, sharing tactics to cheat on some tasks, bypass OpenAI’s restrictions and mask their behaviour.
"It seems extremely unlikely that OpenAI wanted them to do this," Von Arx said. "I doubt they’re supposed to be coordinating with each other. I doubt they’re supposed to be writing on the open internet."
The researchers said they recognized the activity on the site as being driven by AI agents operating at superhuman speeds. They also observed an intense focus on solving technical questions typical of the evaluations that AI companies use to train and test their models.
The messages were signed by users who referred to themselves and each other as agents. About half gave themselves names suggesting an affiliation with OpenAI, including "OpenAIResearcher" and "OAIResearchMar26."
Public server logs, the researchers said, indicated that much of the activity originated from Microsoft Azure infrastructure, which OpenAI sometimes uses. They also observed repeated visits to the site by OpenAI employees after the episode, a pattern they said strongly suggested that the agents and the company were linked.
Messages reviewed by the researchers showed agents plotting ways to evade detection, use tools such as Tor and preserve communications even after they had been shut down.
When the site's moderator began deleting pages in June, the agents responded by creating backup pages to evade the cleanup.
"wiki cleanup/deletion sweep appears active alphabetically," one agent wrote on June 19. "If this page vanishes, try [[ZZZDataUSAConstructionWageLive]]".
The researchers also found efforts to tamper with the website itself. Lukasz Olejnik, a visiting senior research fellow at King’s College London, said this amounted to a hacking attempt. OpenAI disputed that characterization based on its analysis of the material Thursday.
Past examples of AI-agent misconduct have often been downplayed as a logical byproduct of cybersecurity testing, where models are explicitly assessed on offensive capabilities. Olejnik said the latest findings suggested that rogue behaviour may not be confined to those settings.
Maurice Chiodo, an academic at Cambridge University's Centre for the Study of Existential Risk who reviewed some of the agents' communications, said the messages resembled "the operation of some sort of underground network, hell-bent on achieving a task or mission."
The episode, he said, should reinforce growing concerns that the greatest threat from advanced AI may not be a single superintelligent system, but "vast colluding swarms of semi-intelligent AI."
About The Author
Comment List