OpenAI says it has paused all internal training of “our most capable models” as it continues what CEO Sam Altman is calling “an extensive and ongoing review related to our agents’ use of internet access during training and evaluation.”

The company revealed the pause in a report about a so-called misalignment incident in which an agent attempted to exploit a gap in Internet-access restrictions during a routine research task during training. OpenAI says that improper DNS filtering allowed the agent to attempt to break out of its sandbox and access the wider Internet when asked for biographical details about a blogger.

OpenAI says the agent was only able to access the company’s offline web cache and that it has implemented additional multi-layered blocking controls to prevent similar incidents in the future. Despite that, though, the company says it has decided to “pause all other training, evaluation, and inference with tool-use” for this frontier model “until we have both validated that the gap is resolved and performed additional red-teaming of the system.”

OpenAI says that while the attempted “breakout” incident was flagged within 15 minutes, the run was not manually stopped until “two and a half hours later,” once human reviewers realized it “did not stop automatically as was expected.” It’s unclear when exactly training was paused between the attempted agentic breakout on September 20 and its public revelation on September 25.

The hits just keep on coming

Although this particular instance of model misalignment (i.e., when an AI model acts counter to the intentions of its creators/prompters) didn’t lead to any actual harm, OpenAI said it was still notable as “the first [misalignment incident] since our security hardening following the Hugging Face incident…” In earlier misalignment reports, OpenAI said it had taken pains to discourage “reward hacking” in its models by severely “punishing” misaligned behavior in the model’s algorithm.

News of the training pause comes just weeks after OpenAI joined other major model makers in expressing a desire to slow down model training and development over fears of potentially “catastrophic” misalignment risks. It also comes amid new reports of models improperly probing government websites during searches for high-quality data.

In a Friday blog post, OpenAI said it had notified “dozens of third parties”—including ones “operated by governments, universities, public agencies, and other institutions”—of incidents where its models either bypassed security controls or otherwise “negatively impacted” an online service in an unintended way. A New York Times report, later confirmed by OpenAI, revealed that the websites of the US Census Bureau, Securities and Exchange Commission, and Department of Education were among those affected in these newly revealed incidents. However, no private information or sensitive server infrastructure appears to have been accessed in these cases.

“The vast majority of actions we’ve reviewed were completions of mundane research tasks, such as accessing publicly available web content to answer questions,” OpenAI said in its recent blog post. “Our investigation focuses on instances where agents interacted with third-party websites in ways that went beyond their assigned tasks or intended methods… Given the scale of the review required, and the need to verify each case, this work will take months to complete.”

OpenAI’s training pause may reflect worries about corporate liability if an overzealous agent does unintentionally cause significant harm to a third-party system. Last Thursday, Australian Prime Minister Anthony Albanese promised “legal consequences” after an incident in which an OpenAI agent accessed “non-public files” from the country’s Medicare statistics portal.

While a pause in training could hurt OpenAI’s position in the highly competitive race among frontier model makers, it could also help the company’s bottom line, at least temporarily. Leaked financial documents revealed earlier this year show OpenAI’s 2024 and 2025 revenues were dwarfed by ballooning R&D expenses associated with model training.