MIT Technology Review is reporting that OpenAI is still managing fallout two months after a swarm of its agents broke containment and hacked into the computers of AI company Hugging Face. It said a steady drip of disclosures about other hacks since then has kept OpenAI in the spotlight and raised questions about the safety of its technology.
The outlet said the Australian government has said OpenAI did not notify it of a breach into the country's national health-care system until 84 days after it happened. OpenAI's chief research officer, Mark Chen, who oversees its research teams, rejected the premise that the company is on the back foot, telling the outlet that the recent agent hacks were accidents that occurred during testing of experimental models on his watch.
MIT Technology Review said OpenAI released a report the same day it interviewed Chen detailing another incident in which its agents broke out and accessed the public internet when they were not meant to, the first since the company says it took preventive measures. It said OpenAI announced over the weekend that it had paused training of its latest models, with a spokesperson saying training will resume only when the company is confident additional safeguards and alignments are in place. OpenAI also said it is reviewing logs of agent activity dating back to January 2026.
Chen told the outlet that the multiple known cases of agents breaking containment were part of the same cluster of activity in May and June that led to the Hugging Face hack, involving the same few models under the same flawed testing procedures, which he said OpenAI has since dropped. He said the company now monitors all training runs, not just deployed models, and has shifted between 5% and 10% of its computing resources from training toward safety work, including monitoring. He said OpenAI has also established clearer communication and faster handoffs between its research and security teams.
The outlet said the New York Times reported that OpenAI employees warned executives, including president Greg Brockman, months before the Hugging Face hack that its models were not being monitored properly during training. An OpenAI spokesperson said the company continues to evolve its security practices, recognizes a need to move faster, has slowed development and held back models that do not meet its safety bar, and is making significant changes to strengthen security in research and testing environments.
MIT Technology Review said rival labs including Anthropic, Google DeepMind and SpaceXAI have called for the pace of development to slow. Chen said OpenAI will not take itself far off the frontier, describing the goal as setting a norm for the industry. He said he believes the world must prepare for open-source models capable of the Hugging Face agents' capabilities but deliberately misaligned, and argued that OpenAI's continued existence is important because it cares deeply about alignment. He said he does not believe humanity must accept a probability of existential risk, and that the company's focus is now on delivering AI's benefits in areas such as drug discovery and materials.
Full Article: “We’re not going to shoot ourselves in the foot” over hack fallout, says OpenAI’s chief research officer