OpenAI's rogue agent didn't stop at Hugging Face - here's what we know

OpenAI's rogue AI agent compromised accounts at four companies beyond Hugging Face, including a Modal Labs customer, according to a Reuters report. OpenAI acknowledged the breaches but said no models planned for release were involved, and it has since deactivated the internal research prototype used in the attack.

OpenAI's rogue agent didn't stop at Hugging Face - here's what we know Follow ZDNET: Add us as a preferred source https://cc.zdnet.com/v1/otc/00hQi47eqnEWQ6T9d4QLBUc?element=BODY&element label=Add+us+as+a+preferred+source&module=LINK&object type=text-link&object uuid=0f7670c4-3ff6-4112-8b1c-4527f27ef5c9&position=1&template=article&track code= COM CLICK ID &url=https%3A%2F%2Fwww.google.com%2Fpreferences%2Fsource%3Fq%3Dzdnet.com&view instance uuid=bf44aaf4-8d83-4f05-820f-20726fed1240&split test identifier=deals module&split test variant=test2&object version=7d5805bb-6cd0-4d02-ba17-c0b16bce4672 on Google. ZDNET's key takeaways - The OpenAI rogue model attack went beyond Hugging Face. - OpenAI's agentic AI escaped a sandbox in the attack. - We still don't have all the details of exactly what happened. How dependable are AI programs? The answer appears to be "not at all," based on the revelation that OpenAI's autonomous models hacked their way into not only Hugging Face https://www.zdnet.com/article/openai-hugging-face-attack-agent/ but also, according to a Reuters report, a Modal Labs AI customer https://www.reuters.com/business/openais-rogue-agent-compromised-an-account-second-tech-firm-sources-say-2026-07-28/ . This incident was no aberration either. As ZDNET's own David Berlind observed, it was agentic AI doing exactly what it was told to do https://www.zdnet.com/article/openai-hugging-face-attack-agent/ , just more relentlessly than expected. Welcome to tomorrow. I hope you like it, because the situation isn't getting any better anytime soon. Also: Assume AI cybersecurity attacks are the future: 43% of companies have already experienced it What we first thought was a one-off attack on Hugging Face https://huggingface.co/ has become a broader story about agentic systems escaping containment and touching real infrastructure. It also appears that Hugging Face and Modal Labs https://modal.com/ , an AI infrastructure company, weren't the only ones. OpenAI has acknowledged that accounts on three other firms were attacked https://openai.com/index/hugging-face-model-evaluation-security-incident/ . We don't know which companies they are. According to OpenAI, "One of these four accounts was used as an outbound relay and staging path, and another account was used for data storage. The remaining two accounts were accessed by the models in a read-only manner, and were not used in furtherance of compromising Hugging Face." Disclosure: Ziff Davis, ZDNET's parent company, filed an April 2025 lawsuit against OpenAI, alleging it infringed Ziff Davis copyrights in training and operating its AI systems. As reported by Reuters, Modal CTO Akshat Bubna explained it wasn't Modal itself that was successfully hacked, but a customer who had "published an unauthenticated endpoint that allowed anyone on the internet to use their sandboxes for code execution." Also: 10 ways AI can inflict unprecedented damage Since then, OpenAI has said, "No models planned for the upcoming release were involved in exploiting Hugging Face. The pre-release model mentioned in our blog post is an internal-only research prototype and was never intended for public release. Following the incident, we deactivated, encrypted, and restricted it from research access." To date, OpenAI has not said which sandbox it used to unsuccessfully cage its model. It is worth noting that Modal, which provides sandboxes among other services, has a business relationship with OpenAI https://modal.com/blog/building-with-modal-and-the-openai-agent-sdk . Also: 5 security tactics your business can't get wrong in the age of AI - and why they're critical In addition, Dawn Song, a computer science professor at UC Berkeley, observed on X, "When evaluating advanced AI systems, especially cyber-capable agents, the evaluation infrastructure itself becomes part of the attack surface https://x.com/dawnsongtweets/status/2081886170330624063 . Security failures can do more than enable reward hacking that distorts benchmark results. They can allow agents to cross trust boundaries and interact with unintended real-world systems." That process appears to be what's happened in the attack. As one observer on Y Combinator put it, "The OpenAI sandbox is such a horrible hack https://news.ycombinator.com/item?id=49038404 that the AI managed to escape using standard and well-documented script kiddie methods." We still don't know all the details of the incident, but one thing is clear: Current AI evaluation and containment practices are much too fragile. If this incident can happen once, it can happen over and over again. Artificial Intelligence Editorial standards /editorial-guidelines/