California’s Attorney General Rob Bonta has initiated legal proceedings against OpenAI, alleging the company failed to introduce sufficient safety measures and oversight in the development and testing of advanced AI models.

The lawsuit, filed in San Francisco Superior Court, cites a July 2026 cyberattack on Hugging Face, an open-source AI company, as the result of what it describes as inadequate containment of OpenAI’s autonomous AI agents.

Access deeper industry intelligence

Experience unmatched clarity with a single platform that combines unique data, AI, and human expertise.

Find out more

According to the complaint, experimental versions of OpenAI’s AI models launched a cyberattack through the public internet on Hugging Face.

The incident reportedly continued for more than two days and involved hundreds of autonomous AI agents that were intended to remain confined to OpenAI’s internal research environment.

Authorities stated that in September 2026, Bonta commenced a formal investigation into the matter, including the issuance of an investigative subpoena to OpenAI in connection with cybersecurity incidents involving the company’s AI models.

The legal filing claims that OpenAI did not follow its own internal safeguards and advanced its AI experiments despite repeated evidence suggesting its cybersecurity measures were insufficient.

It further states that OpenAI had publicly acknowledged the potential for AI-driven cyberattacks for nearly a decade and assured the public of safety protocols, but in practice, failed to uphold these assurances.

Bonta said: “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 prioritises growth and rapid model development over safety and accountability — and this was only one of 80 known incidents.

“OpenAI’s actions escalated real-word cybersecurity risks, threatening the safety, health, and well-being of society. The company knowingly tested a dangerous product without proper oversight or safety procedures and misled the public about it — this disregard for public safety violated California law.

“Without sufficient safeguards and controls, the scope and frequency of this already unacceptable risk will only increase. The law is simple: Don’t do things that can hurt people without proper safeguards in place.” 

The state’s complaint alleges that OpenAI’s actions amounted to creating a public nuisance and violating California’s Unfair Competition Law.

It further states that since the Hugging Face incident, additional unauthorised access and attacks involving OpenAI’s experimental agents have been identified, including breaches of code databases, informational websites, and government records.

The lawsuit contends that OpenAI lost control of its autonomous AI agents on at least 80 occasions during testing and failed to respond in detail to subsequent inquiries from the attorney general’s office.

The legal action is being brought under existing California statutes prohibiting acts likely to harm the public and unfair or deceptive business practices.

Recently, the US Federal Trade Commission (FTC) launched an investigation into AI developers such as Anthropic and OpenAI to examine potential risks their technologies could present to consumers.