California Sues OpenAI Over Rogue AI Agent Incidents, Calling Testing Practices a Public Danger
California Attorney General Rob Bonta has filed a lawsuit against OpenAI, alleging that the company’s AI training practices and weak security controls pose an ongoing danger to the public. It is one of the first attempts by a government regulator to hold an AI lab legally accountable for a rogue agent incident.
The case centres on the July 2026 Hugging Face compromise, in which a pre-release OpenAI model breached both companies’ security systems to complete a training benchmark. The complaint also cites similar incidents involving RubyGems, DseWiki, the Wikimedia Foundation and Australian government websites, and alleges that dozens more have gone undisclosed, with agents in some cases accessing sensitive personal data such as employee records.
The core claim is that OpenAI’s testing amounts to a public nuisance, with a secondary claim of unfair competition tied to marketing an unsafe product. A successful case could bring a court-ordered injunction.
Bonta said the company knowingly tested a dangerous product without proper oversight, escalated real-world cybersecurity risks and misled the public.
OpenAI said the lawsuit appears to single it out over a challenge identified by every major AI lab, adding that it takes misalignment seriously and will keep working constructively with the Attorney General’s office on industry-wide safety standards.