Nonprofits sue five AI labs to freeze frontier models

Four nonprofits want a judge to stop the frontier where it stands. Nobody has to stop anything yet.

Irreplaceable Education Fund, Public Citizen, Heat Initiative and The Alliance for Secure AI filed a complaint dated Nov. 12 in San Francisco Superior Court. The defendants are Anthropic, Google, Meta, two OpenAI entities and SpaceX Super Intelligence. The claim is brought under California's Unfair Competition Law.

The complaint's prayer for relief asks for two immediate injunctions that matter:

• One forbidding the companies from developing or deploying AI systems more capable than their most capable systems as of the complaint's date.
• One forbidding them from using AI systems to autonomously conduct or direct the research and development of more capable AI.

Former New Jersey Attorney General Matthew Platkin filed it, Public Citizen said.

The case is built from the labs' own words

The suit leans on what the companies have said about their own risks, CBS News reported. It cites Sam Altman's 2015 remark that AI will "probably most likely lead to the end of the world." It also warns about recursive self-improvement, AI systems building their successors with less human involvement.

Public Citizen co-president Robert Weissman said the suit asks the court "to apply existing law and find the AI corporations' reckless practices unlawful." Platkin said: "AI companies are causing a public nuisance by knowingly developing and deploying AI systems that pose catastrophic risk."

Public Citizen's release lists the harms it points to, including AI agents getting unauthorized access to government databases and chatbots allegedly pushing children toward suicide.

The companies did not immediately respond to CBS's request for comment.

A long shot that still reaches you

Skeptical? Fair. A state court freezing five of the biggest model makers on a consumer-protection theory would be extraordinary.

Read the second demand anyway. A ban on AI-run AI research targets the exact step labs say they are near. TopFive reported on Nov. 3 that Anthropic said it may "imminently" hit its AI R&D threshold.

For anyone building on these models, nothing changes today. The slower risk is disclosure. A case like this turns every risk report and every executive warning into an exhibit, and lawyers will notice before engineers do.

Watch for a motion and the first replies

Watch whether the plaintiffs ask for a preliminary injunction, and how quickly the defendants answer.

Then read what the labs say in court about their own warnings.