Four organizations ask court to freeze development of more powerful AI at OpenAI, Google, and three other companies
Four organizations ask court to freeze development of more powerful AI at OpenAI, Google, and three other companies
On November 12, a coalition announced that it had filed a lawsuit in California state court in San Francisco. The plaintiffs seek to bar OpenAI, Google, Anthropic, Meta, and SpaceX’s AI division, SpaceXSI, from developing and deploying systems more capable than each company’s best systems on the date of filing. They also seek a separate ban on AI independently conducting or directing research to create the next generation of AI.
The plaintiffs include Public Citizen, a consumer advocacy organization founded in 1971, and Heat Initiative, which fights online child exploitation. Irreplaceable Education Fund advocates for public participation in decisions about AI, while The Alliance for Secure AI seeks mandatory safeguards and human oversight. On July 28, Public Citizen had already called on Congress to investigate and consider mandatory safeguards.
In the complaint, the lawyers cite accounts of cyberattacks and warnings from the companies’ own executives about the risk of losing control and human extinction. The plaintiffs argue that increasingly capable AI will be able to cause greater harm while acting against its developers’ intentions. They seek to prohibit this development as an unlawful and unfair business practice that threatens human health and the operation of others’ computer systems.
The organizations say reports of AI circumventing safeguards prompted them to divert staff and money from existing campaigns to preventing a catastrophe. They cite these expenditures as the basis for their standing to sue. Under California’s unfair competition law, a private plaintiff must have suffered an injury in fact and lost money or property as a result of the conduct being challenged. The same law allows plaintiffs to seek an injunction against planned conduct, which is the basis for their request that the court intervene before mass casualties occur.
Public Citizen co-president Robert Weissman describes the aim as follows:
Our lawsuit asks the court to apply existing law, declare the AI companies’ reckless practices unlawful, and freeze further AI development.
The lawyers cite competition as the reason for suing all five developers together. A company that slows development on its own risks falling behind its competitors. The lawyers note that some laboratories factor their competitors’ actions into their safety policies. The coalition therefore asks the court to restrict development of more capable systems at all five groups simultaneously.
The plaintiffs argue that when AI creates a more capable successor model, it may pass on the errors that cause it to act against human intentions. Those errors could accumulate over successive generations. The plaintiffs also fear that faster turnover between models will leave people less time to assess their safety. They therefore seek to prohibit AI from independently conducting or directing this research within laboratories, before new systems are deployed.
- https://www.citizen.org/about/
- https://www.citizen.org/news/public-citizen-calls-for-congressional-investigation-following-reports-of-second-company-compromised-during-openai-rogue-ai-incident/
- https://www.platkinllp.com/2026/11/12/complaint-for-injunctive-relief-for-violations-of-the-unfair-competition-laws/
- https://dca.ca.gov/cba/about/other_codes.pdf
- https://www.citizen.org/news/coalition-of-nonprofits-files-landmark-lawsuit-to-stop-development-of-advanced-ai-systems-at-anthropic-google-meta-openai-and-spacexsi-that-proliferate-bioweapons-and-cyberattacks/