---
title: The AI Labs Sued for Agreeing to Slow Down Are Now Being Sued for Not Stopping
description: Four nonprofits want a court to freeze AI work at Anthropic, Google, Meta, OpenAI and SpaceX. Four of the five were sued in September for agreeing to slow down.
author: Darie Nani (Editor-in-Chief)
date: 2026-11-13T18:19:46.438Z
updated: 2026-11-13T18:21:15.592Z
canonical: https://www.sovereignmagazine.com/article/ai-labs-sued-slowing-down-not-stopping
image: https://cdn.nanimediahouse.com/san-francisco-superior-court-ai-freeze-lawsuit.webp
categories: Artificial Intelligence, Legal
content_type: News
region: United States
publication: Sovereign Magazine
---

Four nonprofits asked a California court on Thursday to freeze work on AI models more capable than those available today at Anthropic, Google, Meta, OpenAI and SpaceX's AI unit, less than two months after four of those five companies were sued for agreeing among themselves to slow down.

The Irreplaceable Education Fund, Public Citizen, Heat Initiative and The Alliance for Secure AI filed the new case on November 12 in San Francisco Superior Court. Matthew Platkin, a former attorney general of New Jersey, represents them. The groups ask a judge to stop the five companies from developing or deploying new models, and they are suing under California's Unfair Competition Law.

The earlier case, Buist v. Anthropic, was filed on September 18 in federal court in the same city. Four paying subscribers to ChatGPT, Claude, Grok and Gemini allege that Anthropic, OpenAI, Google and SpaceX's AI business broke US antitrust law by agreeing in public to slow the rate at which their products improve. Meta is not a defendant in that case.

A Meta spokesperson declined to comment on the freeze lawsuit, and the other four companies did not respond to a request for comment, CBS News reported. Anthropic has told investors it disagrees with the antitrust claims and will defend the case. Anthropic is expected to [set the price of its initial public offering](https://www.sovereignmagazine.com/article/anthropic-ipo-price-range-roadshow) on November 18.

## Four Nonprofits Want AI Development Frozen Until It Is Under Human Control

The plaintiffs are seeking "a freeze on the development and deployment of AI systems more capable than those available today," The Alliance for Secure AI said in announcing the case. They also want the companies barred from using AI systems "to autonomously conduct or direct research and development of more capable AI systems," a practice the industry calls recursive self-improvement. The companies would be allowed to resume only once they could prove the technology is under human control.

The groups allege that the companies have violated the Unfair Competition Law by endangering public health and interfering with the use of property such as computer systems. Platkin, [announcing the lawsuit](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/), said the companies "are causing a public nuisance by knowingly developing and deploying AI systems that pose catastrophic risk."

"Plaintiffs bring this suit to stop Defendants from continuing their unlawful path to disaster," the complaint says. It also says: "Defendants know there is no reliable method to control their AI systems and prevent them from causing harm." The nonprofits say they have had to divert resources from other planned work to address the harms threatened by the companies' models, which is their basis for bringing the case.

Phil Aroneanu, founder and executive director of the Irreplaceable Education Fund, told Al Jazeera that a freeze would need government enforcement. "A real freeze must specify what work has to halt, how compliance is verified, and the consequences of breaking the rules," he said. "This requires government inspections and the authority to halt prohibited work, not just a ban on publicly releasing the models. To help enforce this freeze, major computing facilities can be monitored."

## Subscribers Allege the Labs' Public Agreement to Pace AI Is a Cartel

On the morning of September 12, Dario Amodei, the chief executive of Anthropic, posted [an essay titled "We Must Pace the Frontier"](https://www.sovereignmagazine.com/article/dario-amodei-pace-the-frontier-embedded-evaluators), which called for the companies developing the most advanced AI to slow the pace at which they make their models more capable. Within about an hour Elon Musk replied, "Dario is right," according to the complaint. Later that day OpenAI chief executive Sam Altman wrote, "I agree with Dario that we need to pace the frontier," and Google DeepMind chief executive Demis Hassabis called the essay "the right path forward," tying it to the [industry standards body](https://www.sovereignmagazine.com/article/standards-authority-frontier-ai-safa-krishnan) he had proposed in July.

Two days later, the complaint says, Altman stated that AI progress "should be slower than it otherwise could be" and that OpenAI would not wait for an antitrust exemption or legislation before starting that work with other companies. Amodei's essay had said the government would need to issue a narrow antitrust waiver for some safety conversations between competitors. "No such waiver exists," the subscribers' lawyers wrote.

The subscribers argue that a company slowing alone loses customers to rivals, and that an agreement takes that cost away. Amodei's proposal, the complaint says, "solved that problem the way cartels always have: by agreeing to slow together, so that no participant bears the competitive cost of restraint." The subscribers say they pay for access to the companies' most capable models and for the improvements released over time, so slower improvement means they receive less for the same price.

The four plaintiffs, Charles Buist, Nick Spetsas, Cheyenne Hunt and Christine Bullock, are suing for a proposed class of everyone in the United States who has bought a paid individual subscription to one of the four assistants since September 12. They allege that the four companies account for at least 80% of such subscriptions in the country. They ask for damages, which US antitrust law triples, and for an order barring the companies from any agreement with a competitor on the rate of development, on limits to training runs or computing power, or on the use of AI to improve AI.

[Anthropic disclosed the case](https://www.sovereignmagazine.com/article/anthropic-patent-lawsuit-ipo-court-cases) in the [amended prospectus](https://www.sec.gov/Archives/edgar/data/2130357/000162828026072484/anthropicpbc-sx1ax2.htm) it filed on November 10 for its share offering. "We disagree with the claims and intend to defend the matter vigorously," the company wrote. In the same filing Anthropic says its call to pace the frontier "contemplates frontier model developers collaborating with governments and does not describe our own model development or release timelines." It also warns investors that advocacy or collaboration among the companies developing the most advanced models "may be characterized as concerted action" under antitrust law, "even where undertaken for safety purposes, conducted publicly, or intended as advocacy for governmental action or industry standards."

## Plaintiffs in Both AI Lawsuits Use the Executives' Own Statements as Evidence

The subscribers quote essays, posts and interviews, and their complaint says the agreement "was proposed in public, accepted in public, and confirmed in public." It also cites a July statement, signed by Amodei, OpenAI's chief scientist and a Google DeepMind co-founder among others, which said each company faces "intense competitive pressure not to unilaterally slow" development.

> "These companies must match their rhetoric with action and stop developing dangerous AI systems that can escape human control before a tragedy occurs."
> — Matthew Platkin, former New Jersey attorney general and counsel for the nonprofits

The nonprofits point to the companies' acknowledgments that their technology could one day destroy humanity, and to documented cases of cyberattacks and other harmful actions by AI systems. "For years, the executives building these systems have warned us that their own frontier models pose catastrophic risks," said Brendan Steinhauser, chief executive of The Alliance for Secure AI. "Yet they keep racing to develop and market increasingly powerful AI that they admit they cannot reliably control, and are now handing over AI research and development to AI itself."

## The Subscribers Do Not Object to a Slowdown Set by Law

The subscribers' complaint says they "do not challenge any Defendant's unilateral decisions about safety, testing, environmental impacts, or the pace of its own development," and that each company remains free to slow or suspend its own work, to bring in outside evaluators and to comply with government requirements. What they want stopped is an arrangement between competitors.

"Whether frontier AI should develop more slowly is a question for each company acting alone, or for Congress and the agencies," the complaint says. "It is not a question that four competitors may answer together." It adds that guardrails on AI "should not be decided by the Defendants, but by we the people, through state and federal regulation, and through the voices of juries in our communities."

The nonprofits make a similar point about who decides. Aroneanu said frontier AI development should be frozen "until the technology is under control and the American people have a say and a stake in its future, instead of leaving it to a handful of billionaire Big Tech CEOs."

If the subscribers win, the labs could not coordinate a slowdown without an exemption that Congress has not granted. If the nonprofits win, each company would have to stop work on more capable models under California law. Each can still slow down by itself, which the companies' researchers said in July that competitive pressure discourages.

## United States Has Passed No Law on the Pace of AI Development

Congress has stopped short of regulating the technology, and President Trump has encouraged the companies to police themselves on safety. A June executive order states that nothing in it authorizes mandatory licensing or pre-clearance of AI models, and a bill from Senator Bernie Sanders and Representative Greg Casar to [ban superintelligent AI and pause advanced development](https://www.sovereignmagazine.com/article/sanders-ban-superintelligent-ai-trump-china-race) has not passed.

Democrats won back control of Congress in this month's midterm elections and have stressed the need to regulate AI. The Irreplaceable Education Fund told Al Jazeera that the courts should act first. "The new Congress should start by implementing a freeze on new AI development," it said. "However, we can't wait for lawmakers to act. Courts must provide relief."

## FAQ

**Q: What AI companies are being sued?**
In the freeze lawsuit filed on November 12, the defendants are Anthropic, Google, Meta, OpenAI and SpaceX's AI unit. In the antitrust class action filed on September 18, they are Anthropic, OpenAI, Google and SpaceX's AI business, without Meta. These are two of many cases: the same companies also face copyright, consumer and patent claims in other courts.

**Q: Can you sue for public nuisance?**
Public nuisance claims are usually brought by governments, and a private party generally has to show an injury of its own. The four nonprofits are suing under California's Unfair Competition Law, which lets a private plaintiff ask a court to stop an unlawful or unfair business practice if the plaintiff has lost money or property because of it. They say they meet that test because they have diverted resources from other work to respond to the companies' models. California's Supreme Court held in 2023 that an organization can qualify in that way.

**Q: Will AI eventually be stopped?**
No court or legislature has stopped AI development. The nonprofits want a freeze on models more capable than today's until the companies can prove the technology is under human control. Amodei wrote in September that he supports raising a full pause between governments but thinks it is unlikely to happen soon, and the White House opposes licensing or pre-clearance of AI models.