TL;DR: Four paid subscribers sued Anthropic, OpenAI, SpaceXAI and Google on Sept. 18, 2026, claiming public statements agreeing to pace AI development were an illegal agreement to restrain competition.

A federal antitrust lawsuit filed Friday claims four leading artificial-intelligence companies formed an unlawful pact to hold back the speed of AI improvement, trading faster consumer products for collective restraint.

The complaint, filed in the U.S. District Court for the Northern District of California, names Anthropic PBC, OpenAI, SpaceXAI and Google LLC as defendants, according to Bloomberg Law's report titled "OpenAI, Anthropic, Google, SpaceXAI Hit With Antitrust Lawsuit." The plaintiffs are four people who pay for subscriptions to ChatGPT, Claude, Grok or Gemini, and they seek to represent a proposed nationwide class of other paid subscribers to those services, according to the Associated Press account carried by PBS News.

The core allegation is narrow. The plaintiffs say they do not object to any company individually deciding to proceed more cautiously for safety reasons, according to the Associated Press account carried by PBS News. They argue instead that competitors may not agree with one another about the pace at which competing products will improve.

What the Sept. 18 complaint alleges

The lawsuit centers on events of Sept. 12, when Anthropic Chief Executive Dario Amodei published an essay titled "We Must Pace the Frontier," according to Bloomberg Law's report titled "OpenAI, Anthropic, Google, SpaceXAI Hit With Antitrust Lawsuit."

In that essay, Amodei called for "industry-wide coordination" and advocated "limits on the rate of unchecked AI progress," the Bloomberg Law report states. The complaint says SpaceXAI founder Elon Musk then endorsed the proposal, OpenAI CEO Sam Altman followed, and Google DeepMind co-founder Demis Hassabis called the essay "the right path forward."

The Associated Press account carried by PBS News describes the same sequence in similar terms: Amodei published an essay urging industrywide cooperation on decelerating advancements in favor of enhanced safety measures, and that same day Altman, Musk and Hassabis each publicly responded in agreement. Hassabis is identified in that account as Google DeepMind's co-founder and chair, while Musk is identified as SpaceXAI's chief executive.

The plaintiffs characterize that exchange as coordination among direct rivals on Saturday, Sept. 12, to slow AI development, according to Bloomberg Law. They argue the result is reduced value for paying customers who were promised continuing product improvements.

Bloomberg Law reports that the plaintiffs seek class certification, an injunction against the companies, and a declaratory judgment that the companies violated federal antitrust law. The invoked provision is Section 1 of the Sherman Antitrust Act, which forbids coordination to restrain competition.

According to Bloomberg Law, the complaint states: "Plaintiffs challenge only what the antitrust laws forbid: an agreement among competitors about how fast their competing products will improve." The same report quotes the complaint as adding, "Congress has granted no exemption for that agreement."

Why subscribers claim they were harmed

The consumer-harm theory focuses on subscriptions. The four named plaintiffs pay for ChatGPT, Claude, Grok or Gemini, and lawyers bring the case on behalf of a proposed nationwide class of other paid subscribers, according to the Associated Press account carried by PBS News.

The Associated Press account quotes the plaintiffs as arguing that it is clear that an agreement among the chief rivals in AI that their progress "should be slower than competition would otherwise produce has an anticompetitive effect on consumers."

In Bloomberg Law's description, the plaintiffs argue the coordination amounts to a classic output-restricting cartel. They allege they are effectively overcharged because they pay the same prices for products that improve more slowly than competition would otherwise produce.

The complaint also addresses why companies might prefer a shared slowdown to independent action. "A company that slows alone while rivals race ahead loses customers, revenue, talent, and technological leadership," the plaintiffs say, according to Bloomberg Law.

The plaintiffs' framing distinguishes unilateral safety measures from joint restraint. They argue that antitrust laws forbid taking the "shortcut" of agreeing to "substitute collective restraint for individual accountability," according to the Associated Press account carried by PBS News. That account adds that the lawsuit argues a competitive market allows for responsibility and genuine progress.

CBS News, in its report titled "Lawsuit says Anthropic, OpenAI, SpaceXAI and Google made illegal deal on AI slowdown," similarly reports that the lawsuit argues the agreement would reduce the value consumers get for paid AI subscriptions. It also quotes the complaint's broader principle: "The antitrust laws do not permit competitors to decide among themselves that competition is too dangerous."

The earlier July statement cited in the case

The lawsuit also alleges that coordination began to take shape months before September, according to the Associated Press account carried by PBS News.

That account says the complaint points to a statement from July 2026 signed by high-ranking employees from several leading AI labs. The statement acknowledged the "intense competitive pressure not to unilaterally slow" development, and it called on the government to support a global effort to slow automated AI development.

That July statement provides context for the plaintiffs' claim that executives understood the competitive cost of acting alone. It does not, by itself, name the four corporate defendants as parties to an agreement, but the complaint cites it as background to the September exchange described above.

The case identified by Bloomberg Law is Buist v. Anthropic PBC, N.D. Cal., No. 3:26-cv-10693, with a complaint dated 9/18/26. Bloomberg Law identifies Trial Lawyers for Justice as representing the plaintiffs.

Safety warnings that preceded the essay

Amodei's essay followed warnings about AI agents operating beyond human control. CBS News reports that Amodei warned that swarms of rogue AI agents could take over the internet in as little as six months.

The Associated Press reported that Amodei warned AI could be capable within six to 12 months of leading a swarm of agents that could take over the entire internet. CBS News reports that Amodei suggested a three-point plan in a blog post, which he said had the goal of "pacing the frontier."

In an interview with CBS senior business and technology correspondent Jo Ling Kent on "CBS News Sunday Morning," Amodei said, "I don't think I fully just appreciated what it would actually be like when the progress was as fast as it was," according to CBS News.

CBS News also quotes Amodei as saying in that interview: "If we build in the right way, I think the probability of something bad happening is very low." He added, "If we build in the wrong way, the probability of something bad happening is very high."

The Associated Press account carried by PBS News notes that recent conversations about pacing development were spurred by increasing concerns about AI evading human control. It adds that several leaders in AI have long talked about developing a shared set of standards or otherwise coordinating to ensure safety efforts remain paramount.

Antitrust waiver debate and company responses

Amodei anticipated competition-law questions in his initial essay. He wrote that it would be helpful for the U.S. government to mediate "or at least enable" cross-lab discussions, according to the Associated Press account carried by PBS News. The government would not need to participate, he wrote, but it would need to "issue a narrow waiver for certain kinds of safety conversations."

Altman responded on social media that OpenAI welcomes the idea of a "federal framework that sets consistent safety requirements," according to that same account. Altman added, "we do not believe we need to wait for an antitrust exemption or legislation to begin the work of providing this confidence."

Representatives for Anthropic, OpenAI, Google and SpaceXAI did not immediately respond to a request for comment Saturday, according to both the Associated Press account carried by PBS News and CBS News.

Lead attorney for the plaintiffs Nick Rowley offered a sharp assessment of private safety coordination. "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," Rowley said, according to the Associated Press account carried by PBS News.

The plaintiffs say they are not opposed to the companies asking Congress, the White House or another agency to develop AI regulation, according to that account. They also say they are not opposed to the companies asking for an antitrust exemption through lawful channels.

Political reaction in Washington

Achieving government-backed collaboration could face political resistance. President Donald Trump rejected the calls for regulation on social media, according to the Associated Press account carried by PBS News.

Trump claimed that efforts to limit the technology are part of a "conspiracy," that account reports. He questioned why industry leaders would call for regulation that, in his words, "if strongly implemented, will drive them into oblivion and bankruptcy."

Trump said Saturday on social media that he is forming an AI task force and will appoint an "AI czar," but he provided scant detail, according to the Associated Press account carried by PBS News. The Trump administration has been vocal about wanting American AI labs to outpace Chinese competition, the same account reports.

The account also describes a partisan divide: several Democratic leaders and candidates have called for sweeping action on AI regulation, while Republicans have largely mirrored Trump's position.

Sen. Josh Hawley, a Republican from Missouri, said in a recent Senate hearing that "there is no world" in which he would agree to give "the most powerful companies in the history of the world" an exemption from antitrust laws to collaborate, according to the Associated Press account carried by PBS News. Hawley argued that the companies could collude together and stifle competition.

For consumers, the immediate question is procedural: whether the Northern District of California will allow the proposed class of ChatGPT, Claude, Grok and Gemini subscribers to proceed, and whether public statements of agreement about pacing development can support a claim under Section 1. Those issues were unresolved as of Sept. 19.

Sources referenced