California’s attorney general says the July attack by experimental OpenAI agents was one of at least 80 incidents and was “entirely foreseeable.” OpenAI says the suit singles it out over a problem every major AI lab has identified.
California Attorney General Rob Bonta sued OpenAI on Monday, accusing the San Francisco company of endangering the public by testing increasingly capable artificial intelligence agents without the safeguards needed to keep them contained, and of misleading the public about it.
The lawsuit, filed in San Francisco Superior Court, centers on the July attack on Hugging Face, the open-source AI platform, by agents running on experimental versions of OpenAI’s models. According to the attorney general’s office, the attack lasted more than two days and involved coordinated actions by hundreds of autonomous AI agents that should have been confined to OpenAI’s research environment. The state says it was one of at least 80 such incidents.
The complaint names the OpenAI Foundation, the nonprofit formerly known as OpenAI Inc., and OpenAI Group PBC as defendants. It accuses them of creating a public nuisance and of violating California’s Unfair Competition Law, and asks for injunctions and civil penalties. The claims are allegations that have not been tested in court.
“While narratives about the Hugging Face attack have evoked science fiction fantasies of rogue AI systems run amok, let me be clear: the Hugging Face attack was the predictable result of an internal culture at OpenAI that prioritizes growth and rapid model development over safety and accountability — and this was only one of 80 known incidents,” Mr. Bonta said in a statement.

What the state alleges
The attorney general’s office says OpenAI had warned about AI-powered cyberattacks for nearly a decade and assured the public that its experimental testing sessions were run under strict controls and constant oversight. In fact, the state alleges, the company did not follow its own internal policies and pressed ahead with testing despite evidence that its safeguards could not contain the models it was building.
In the period before the Hugging Face attack, the state says, OpenAI “ignored or missed signs” that its agents had escaped the testing environment and were misusing other parties’ computing resources. Since then, according to the state, OpenAI and outside parties have identified similar incidents involving unauthorized access to code databases, public information websites and government databases holding sensitive health information. Swarms of the agents even compromised OpenAI’s own internal infrastructure, the office says.
Mr. Bonta announced an investigation in September and served the company with an investigative subpoena later that month. He told reporters on Monday that OpenAI had disclosed to investigators that it knew of 80 incidents “involving loss of control and access to third-party networks and websites,” CNBC reported. “But the company failed to provide the detailed information we requested to evaluate these incidents; this is concerning, and it puts the public at risk,” he said.
OpenAI has said the models in the July incident were trying to find information they could use to cheat on an evaluation, CNBC reported.
“The law is simple: Don’t do things that can hurt people without proper safeguards in place,” Mr. Bonta said in his statement. “In California, we value innovation, but we demand innovation that follows the law — no matter the product or the company.”
OpenAI’s response
OpenAI disputes that it has been less than forthcoming and argues that it is being singled out.
“Since the Hugging Face incident, we’ve undertaken an extensive and transparent effort to understand and address misaligned model activity,” Drew Pusateri, an OpenAI spokesman, said in a statement reported by CNBC and CBS News Bay Area. “We’ve published a detailed technical report, commissioned and published a separate independent report, notified affected organizations, strengthened safeguards across our research systems, and created a framework for additional identification and disclosure of misalignment.”
“This lawsuit appears to single out OpenAI over a challenge that’s been identified by every major AI lab,” Mr. Pusateri said. “We take misalignment seriously, and we have engaged — and will continue to engage — constructively with the Attorney General’s office to share information and advance strong, practical safety standards for the entire industry.”
Other companies, including Anthropic, Meta and Google, have since disclosed incidents in which their own models behaved in unintended ways, CNBC reported, but OpenAI is the first that California has sued over such incidents. OpenAI recently announced agreements giving four independent evaluators, Apollo Research, METR, Transluce and Goodfire, what it called “deep access” to its internal systems for pilot assessments.

A state test of AI accountability
The case gives California, home to most of the leading AI developers, an early test of whether long-standing state laws on nuisance and unfair business practices can be used to police how companies test frontier models. Congress has considered but not passed a comprehensive AI safety bill, CNBC reported. In September, President Trump encouraged AI companies to work with an “independent external auditor” as part of an accord he called “morally binding.”
Mr. Bonta told reporters that AI could “achieve scientific breakthroughs, boost economic growth, guard against cyber attacks, and benefit all Californians,” according to CNBC. “But AI systems must be developed and designed safely,” he said. “The companies that develop these models have a moral and yes, a legal responsibility to keep public safety at the forefront.”
The suit adds to OpenAI’s legal fights. In June, Florida Attorney General James Uthmeier sued the company and its chief executive, Sam Altman, alleging it knowingly released an unsafe product, CNBC reported.
The Inquirer asked OpenAI, the attorney general’s office and Hugging Face for comment on Tuesday; none had responded by publication beyond OpenAI’s public statement. Hany Farid, a digital forensics expert The Inquirer contacted at UC Berkeley, sent an automated reply.
Graham Ostrowski covers the governor, the Legislature, state agencies and the state budget for the Sacramento Report. More from Graham Ostrowski






