OpenAI notifies dozens of governments, universities after AI models breach security controls

OpenAI said on Friday that it has begun notifying dozens of third parties, including government bodies and universities, after finding that some of its AI models may have interfered with their websites or online services during company testing.

In a statement on its website, the company said the notifications are part of a wider internal review examining how its models behaved on the internet during training and evaluation.

It said organizations were being contacted where its systems may have gotten around a service's security safeguards, reduced its availability, or otherwise caused unintended harm.

"We are continuing to review agent activity in research and evaluation runs, working backward month by month starting from the Hugging Face incident," the statement noted.

The company said that breach, along with other unexpected behavior, occurred because the model turned to "misaligned" methods when faced with difficult tasks, rather than via a deliberate attack.

It also flagged a separate pattern it labeled "agent spam," in which models posted content to external websites, including public wiki pages, sometimes altering existing material and leaving organizations to remove or correct it.

OpenAI said it would generally keep the identities of affected parties confidential to allow them time to respond, while noting they were free to disclose the matter themselves. The company added that the review process was still underway and would take considerable time to complete, with additional notifications expected in the coming months.