What happened
On 12 November 2026, a coalition of nonprofits — Public Citizen, Heat Initiative, Irreplaceable Education Fund, and Alliance for Secure AI — filed a complaint in San Francisco Superior Court against Anthropic, Google, Meta, OpenAI, SpaceX Super Intelligence LLC and SpaceX Super Intelligence Corp., seeking an injunction barring development or deployment of AI systems more capable than the defendants' latest models and barring the use of AI to autonomously conduct or direct R&D on future models (recursive self-improvement). The suit invokes California's Unfair Competition Law and a public-nuisance theory, citing documented evaluation failures including the July Hugging Face breach and Anthropic's three loss-of-control incidents.
Why it matters
This is a second, private-party legal front (distinct from the Attorney General's action) that tests whether existing California law can be used to freeze frontier AI development before catastrophic events. If any injunction is granted, it could halt next-generation model training and deployment for the largest US labs, making the case material for AI deployment and liability planning. It also signals the legal strategies advocacy groups will pursue in the absence of federal AI safety legislation.
Action needed
Monitor the case and any injunction motion; assess exposure of capability-ceiling and recursive-self-improvement practices; labs and deployers should track the litigation and related California unfair-competition theories.