Four paid AI subscribers filed a civil antitrust lawsuit on September 18 against Anthropic, OpenAI, Google, and SpaceXAI (formerly xAI), in the U.S. District Court for the Northern District of California. The suit alleges the four companies illegally coordinated to slow the pace of their AI development, reducing the value customers get from paid AI subscriptions, according to a CNN report.

According to the report, the lawsuit centers on September 12, when Anthropic CEO Dario Amodei published an essay calling for industry-wide coordination to slow AI development in favor of stronger safety measures. 

That same day, SpaceXAI’s Elon Musk responded, “Dario is right,” and OpenAI’s Sam Altman said he agreed with Amodei’s proposal. Google DeepMind’s Demis Hassabis also expressed support.

A Narrow but Pointed Legal Argument

The CNN report says the plaintiffs aren’t arguing that any single company slowing its own AI development is illegal. 

Their argument is narrower: that competitors publicly coordinating to slow down together violates antitrust law, designed to keep markets competitive by preventing companies from illegally coordinating with each other or abusing market power in ways that hurt consumers or competition.

The plaintiffs argue that the antitrust law stops the AI vendors from taking the “shortcut” that lets companies “substitute collective restraint for individual accountability.”

The lawsuit argues that a competitive market allows for responsibility and genuine progress.

“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 lead attorney for the plaintiffs. 

Representatives for all four companies did not immediately respond to requests for comment.

Amodei himself had flagged this exact risk in his original essay.

He wrote: “For antitrust reasons, it’s helpful for the US government to mediate or at least enable these discussions — they don’t need to participate, but do need to issue a narrow waiver for certain kinds of safety conversations.”  

Republican Senator Josh Hawley rejected that idea directly during a Senate hearing with FBI Director Kash Patel on September 15, days before the lawsuit was filed, according to The Hill.

“There is absolutely no world in which I will consent to giving the most powerful companies in the history of the world — a small group of three or four of them — antitrust exemptions so they can what … collude together?” Hawley said. “Absolutely no way that’s happening.”

Coordination on Safety Now Carries Legal Risk of Its Own

For chief data officers (CDOs) evaluating AI vendors, this lawsuit adds more complication to an already complicated picture. 

Much of the AI vendor governance conversation this year has centered on getting companies to be more transparent and coordinate with each other and with customers. This lawsuit argues that the same kind of public coordination could itself be a legal liability.

If courts find that AI companies coordinating publicly on safety exposes them to antitrust risk, vendors may become more cautious about the very kind of cooperation, shared disclosure frameworks, and joint safety standards that CDOs have been pushing for as part of broader AI governance work.