Anthropic, OpenAI, SpaceXAI and Google are accused in a new lawsuit of reaching an unlawful agreement to slow the development of their artificial intelligence systems.
Filed Friday in the U.S. District Court for the Northern District of California, the complaint alleges that the major AI companies violated antitrust law by coordinating efforts to reduce the pace of innovation. The plaintiffs say the alleged agreement could diminish the value consumers receive from paid AI subscriptions.
According to the lawsuit, the collaboration became public on Sept. 12, after Anthropic CEO Dario Amodei published an essay calling for industrywide cooperation to slow technological advances in favor of stronger safety protections. OpenAI CEO Sam Altman, SpaceXAI CEO Elon Musk and Demis Hassabis, co-founder and chair of Google DeepMind, each publicly expressed agreement with Amodei’s proposal that day.
The complaint says the groundwork for such coordination was laid months earlier. It cites a July 2026 statement signed by senior employees at several prominent AI laboratories, which recognized the “intense competitive pressure not to unilaterally slow” development. The statement urged the government to back a worldwide effort to decelerate automated AI progress.
The plaintiffs contend that an agreement among AI competitors to make progress “slower than competition would otherwise produce” necessarily harms consumers and has an anticompetitive impact.
Four named plaintiffs who subscribe to ChatGPT, Claude, Grok or Gemini are pursuing the case on behalf of a proposed nationwide class of other customers who pay for those AI services.
The customers say they are not challenging any company’s independent decision to prioritize safety by slowing development. Instead, they argue that antitrust rules prohibit the companies from taking a “shortcut” by agreeing to “substitute collective restraint for individual accountability.” In their view, competition is essential to preserving both accountability and meaningful innovation.
“AI will quickly spin out of human control and could kill us all if we allow AI safety and protocol … to be controlled by private self-serving agreements between the world’s most powerful ‘for profit’ technology companies,” said Nick Rowley, the plaintiffs’ lead attorney.
Representatives of Anthropic, OpenAI, Google and SpaceXAI did not immediately respond to a request for comment Saturday.
Amodei addressed the potential antitrust implications in his original essay, saying the U.S. government could mediate—or at least facilitate—talks among AI laboratories. He wrote that officials would not necessarily have to join the discussions, but might need to provide “a narrow waiver for certain kinds of safety conversations.”
Altman responded on social media that OpenAI supports a “federal framework that sets consistent safety requirements.” However, he added, “we do not believe we need to wait for an antitrust exemption or legislation to begin the work of providing this confidence.”
The latest debate over slowing AI development has been driven by growing fears that advanced systems could evade human control. Still, AI executives and researchers have for years discussed common standards and other forms of cooperation intended to keep safety at the center of development.
The plaintiffs emphasize that they are not opposing efforts by the companies to ask Congress, the White House or federal agencies to establish AI regulations. They also say they do not object to the companies seeking an antitrust exemption.
Even so, securing federal support for that type of cooperation may prove difficult.
President Donald Trump has rejected calls for regulation in social media posts, describing efforts to restrict the technology as part of a “conspiracy.” He has questioned why industry leaders would seek rules that, he said, “if strongly implemented, will drive them into oblivion and bankruptcy.” Trump said Saturday that he is creating an AI task force and plans to appoint an “AI czar,” though he offered few details.
The Trump administration has repeatedly stressed the need for U.S. AI companies to surpass Chinese competitors. Although several Democratic leaders and candidates have advocated sweeping AI regulation, most Republicans have taken positions similar to Trump’s.
Sen. Josh Hawley, a Missouri Republican, said at a recent Senate hearing that “there is no world” in which he would support granting “the most powerful companies in the history of the world” an antitrust exemption to collaborate. He warned that the companies could use such an arrangement to collude and suppress competition.
Copyright © 2026 by The Associated Press. All Rights Reserved.
