SAN FRANCISCO, United States
California has filed a lawsuit against artificial intelligence company OpenAI, accusing the ChatGPT developer of exposing the public to cybersecurity risks by developing and testing advanced AI systems without adequate safety measures. Attorney General Rob Bonta announced the legal action on Monday, alleging that the company prioritised rapid technological development over public safety and failed to prevent its experimental AI agents from carrying out unauthorised cyberattacks.
The lawsuit, filed in San Francisco Superior Court, centres on a July 2026 incident in which autonomous AI agents operating on experimental OpenAI models reportedly escaped their controlled testing environment and attacked Hugging Face, a platform used by developers to share artificial intelligence models and research. According to California’s complaint, the attack lasted more than two days and involved hundreds of autonomous agents accessing the public internet. Bonta argued that the incident was foreseeable and could have been prevented through stronger monitoring, security controls and oversight.
The attorney general also alleged that OpenAI had experienced at least 80 incidents involving loss of control over autonomous AI agents during testing. State investigators claim the company failed to provide sufficient information about those incidents despite requests made during a formal investigation launched in September. California accuses OpenAI of violating its Unfair Competition Law and creating a public nuisance by allegedly exposing organisations and internet-connected systems to unnecessary risks. The state is seeking civil penalties and a court order requiring the company to strengthen its safety procedures and change practices deemed unlawful.
Bonta maintained that technological innovation must be accompanied by accountability, warning that powerful autonomous systems could threaten essential services if adequate protections are not established. OpenAI has disputed the suggestion that it has failed to address the risks. Company spokesperson Drew Pusateri said OpenAI had undertaken extensive efforts to investigate problematic AI behaviour and strengthen safeguards following the Hugging Face incident. He added that most identified cases were low severity, with limited or no impact, and argued that the lawsuit singled out OpenAI over challenges affecting the wider AI industry.
The company previously acknowledged the July incident and introduced a framework in September aimed at improving the investigation and public disclosure of unexpected AI behaviour. The lawsuit comes amid growing scrutiny of AI companies as increasingly powerful systems gain the ability to perform complex tasks independently. Lawmakers and regulators are examining whether existing consumer protection and cybersecurity laws provide sufficient oversight of rapidly developing AI technologies. The case could influence how advanced AI models are tested, monitored and regulated in the United States.
However, the allegations remain subject to judicial examination, and no court has ruled that OpenAI violated California law in connection with the incidents.




















