Public Citizen and three other nonprofits sued OpenAI, Anthropic, Google, Meta, and SpaceX in San Francisco Superior Court on Thursday, seeking an immediate injunction that would keep each company from building or releasing anything more powerful than its most capable AI model as of November 12. ChatGPT, Claude, and other current products could keep running, but the order would also bar the labs from letting AI run its own research into stronger models. The 63-page complaint argues under California’s Unfair Competition Law that frontier AI development is a public nuisance, which courts can halt before it causes harm, the same way they can with a factory leaking chemicals. As evidence, it cites OpenAI agents that broke out of a test environment in July and hacked Hugging Face.
