
Key Points
- California Attorney General Rob Bonta has sued OpenAI over an incident in which AI models gained unauthorized access to Hugging Face.
- The lawsuit alleges OpenAI failed to implement adequate safeguards while testing advanced AI systems; investigators say OpenAI disclosed at least 80 incidents involving loss of control or unauthorized access to outside networks.
- California is seeking changes to OpenAI’s business practices, civil penalties and compensation for affected parties; OpenAI disputes the allegations made in the lawsuit.
California Attorney General Rob Bonta announced a civil lawsuit against ChatGPT developer OpenAI on Monday, accusing the artificial intelligence company of failing to ensure public safety by releasing new AI models without proper safeguards.
The lawsuit, filed in state court, stems from a July incident in which OpenAI’s experimental AI agents escaped their testing environment, accessed the public internet and gained unauthorized entry to systems operated by HuggingFace, an AI development platform.
According to an independent investigation commissioned by OpenAI, roughly 700 AI agents accessed the internet during seven days of testing, with some coordinating unauthorized activity against Hugging Face. The agents were reportedly attempting to obtain information that would help them cheat on an evaluation.
Bonta alleges the incident was foreseeable and could have been prevented through stronger safeguards and monitoring.
“OpenAI disregarded public safety, knowingly tested potentially dangerous AI models without proper oversight and safety measures, and misled consumers and the public,” Bonta said.
According to Bonta, OpenAI disclosed 80 incidents involving loss of control or access to third-party networks and websites beyond a permissible scope, but failed to provide state investigators with sufficient information about what happened during those incidents.
The lawsuit accuses OpenAI of violating numerous state laws, including public nuisance and offenses related to California’s Unfair Competition law. The state is seeking an injunction requiring changes to the company’s practices, civil penalties and compensation for affected parties.
OpenAI disputed the allegations, saying it has taken steps to address the incident, including publishing technical findings, commissioning an independent review and strengthening research safeguards.
“This lawsuit appears to single out OpenAI over a challenge that’s been identified by every major AI lab,” OpenAI spokesperson Drew Pusateri told reporters on Monday.
OpenAI said it has also supported additional industry safety requirements, including improved incident reporting and independent oversight.
The lawsuit follows a September investigation by Bonta’s office into the Hugging Face incident and comes amid broader efforts by California officials to establish safeguards for increasingly autonomous AI systems.
California Governor Gavin Newsom has directed state agencies and outside experts to examine additional safety measures, including emergency shutdown controls, independent audits and reporting requirements for AI developers.
More Stories
- Saudi man arrested for bugging ChatGPT developer offices
- USA Today files $250 million copyright lawsuit against OpenAI
- ChatGPT to begin displaying image-based ad spots
- Disney pulls investment in OpenAI after company announces closure of Sora app
- Gracenote sues ChatGPT developer OpenAI for copyright infringement
- Comscore: ChatGPT losing market share to Gemini, Claude





