A federal lawsuit filed Friday accuses four of the world's biggest artificial intelligence companies — Anthropic, OpenAI, SpaceXAI and Google — of striking an illegal deal to slow down the pace of AI development, a move the suit claims cheated paying customers out of the value of their subscriptions. The case, brought on behalf of four named plaintiffs who pay for ChatGPT, Claude, Grok or Gemini, was filed in U.S. District Court for the Northern District of California.
The complaint alleges the companies violated antitrust laws by coordinating efforts to pace their AI systems, turning what the industry has framed as a safety conversation into what plaintiffs call an illegal business agreement between competitors, according to Politico. The named plaintiffs — Cheyenne Hunt, along with Florida attorneys Charles Buist and Nick Spetsas, and California resident Christine Bullock — are seeking class-action status on behalf of other paid subscribers to ChatGPT, Claude, Grok and Gemini who they say were similarly affected, as reported by economictimes.indiatimes.com.
At the heart of the case is the argument that a coordinated AI slowdown would reduce the value consumers get for paid AI subscriptions, per PBS NewsHour. Notably, the plaintiffs say they do not oppose the companies seeking AI regulation from Congress, the White House or another federal agency, nor do they object to the firms pursuing a formal antitrust exemption — their objection is to informal coordination they say happened outside any such process, according to reporting cited in the original Associated Press account carried by OregonLive.
The Essay That Triggered the Suit
The lawsuit traces back to an essay Anthropic CEO and co-founder Dario Amodei published on September 12, urging industrywide cooperation to decelerate AI advancement. As detailed by Reuters, Amodei called on AI companies to slow the rate at which they advance model capabilities and floated a three-step framework built around embedded independent evaluators, coordination among frontier labs on safety standards, and international cooperation on AI risks.
Amodei said he was not calling for a halt to model training or technical progress, but rather urging companies to take adequate time to safeguard their models, the same Reuters report noted. He also proposed that the U.S. government mediate or enable cross-lab safety discussions, and separately suggested the government issue a narrow waiver for certain safety conversations. Reuters also reported that Amodei said such an approach would likely require targeted antitrust exemptions in the United States — a point that became central to the dispute over how companies should legally coordinate.
OpenAI CEO Sam Altman, SpaceXAI CEO Elon Musk and Google DeepMind co-founder and chair Demis Hassabis each publicly responded to Amodei's proposals in agreement, according to PBS NewsHour. But Altman drew a different line than Amodei on the legal mechanics: he said OpenAI does not need to wait for an antitrust exemption or legislation to begin safety work, while also saying the company welcomes a federal framework setting consistent safety requirements, according to the Advocate. Anthropic separately disclosed that a threat-intelligence report found several actors had used its Claude AI models for activities ranging from weapons development and cyber operations to surveillance and fraud, per Reuters.
Companies Stay Quiet as Political Fight Heats Up
Anthropic, OpenAI, SpaceXAI and Google did not immediately return requests for comment, Politico reported — and by Saturday, none of the four companies had responded, according to the AP account republished by OregonLive. The silence has left the legal questions raised by Amodei's proposal, and the lawsuit that followed it, unresolved even as political reaction builds in Washington.
President Donald Trump rejected calls for AI regulation, calling efforts to limit AI a conspiracy and saying strongly implemented AI rules would drive companies into oblivion and bankruptcy, per the AP report carried by OregonLive. Trump has instead announced plans to form an AI task force and appoint an AI czar, consistent with what the Advocate described as the administration's broader light-touch approach aimed at helping American AI labs outpace Chinese competitors.
Not every Republican is on board with letting companies sort out safety coordination themselves. Senator Josh Hawley argued that AI-company collaboration could enable collusion and stifle competition, and said he would not agree to give major AI companies an antitrust exemption to collaborate, according to the AP account. That skepticism echoed Cohere CEO Aidan Gomez, who told the Advocate that the rules for AI cannot be written by a small group of commercially aligned companies behind an antitrust waiver.
Safety Advocates Warn Against Private Deals
Attorney Nick Rowley, one of the plaintiffs' lawyers, warned that AI could quickly spin out of human control and kill people if safety protocols are controlled by private agreements rather than public oversight, according to the AP report. That argument runs parallel to broader doubts about whether an industry-led slowdown is even achievable: Oxford Internet Institute professor Sandra Wachter told the Advocate that a coordinated AI slowdown is theoretically possible but highly unrealistic, since it would require extensive international coordination that no single agreement among U.S. companies could deliver.
The case leaves competing versions of how AI companies should legally coordinate on safety unresolved. Amodei has said an antitrust exemption would likely be necessary for real cross-lab cooperation, while Altman has maintained OpenAI can pursue safety work without waiting on Congress or a formal waiver. With the lawsuit now pending in federal court and no response yet from any of the four companies, that legal question is set to be argued well beyond the opinion pages where it started.
