The Pulse
Four AI Companies Face Antitrust Suit Over Slowdown Calls
Subscribers sued Anthropic, OpenAI, Google and SpaceX in federal court, alleging the companies coordinated to slow improvements to competing AI products. The complaint targets public calls by Dario Amodei, Elon Musk, Sam Altman and Demis Ha
AI.info Team ·
Four companies, one alleged agreement
Four major AI companies now face an antitrust lawsuit over public calls to slow the development of frontier models. Subscribers to Anthropic, OpenAI, Google and SpaceX filed the complaint in federal court in San Francisco on September 18, alleging that the companies coordinated to limit how quickly their competing products improve.
The complaint names Anthropic PBC, OpenAI Inc., Space Exploration Technologies Corp. and Google LLC, which is owned by Alphabet. Plaintiffs say executives’ public statements about coordinating AI safety and development amount to an agreement among competitors, rather than a series of independent views about a dangerous technology.
Trial Lawyers for Justice filed the case on behalf of the subscribers. The plaintiffs seek monetary damages, preliminary and permanent injunctions, and a declaration that the alleged coordination violates federal antitrust law.
Dario Amodei’s essay becomes the lawsuit’s starting point
The complaint focuses on a September 12 essay by Anthropic Chief Executive Dario Amodei titled We Must Pace the Frontier. According to the filing, Amodei called on leading AI laboratories to work together to impose “limits on the rate of unchecked AI progress.”
SpaceX founder Elon Musk, OpenAI Chief Executive Sam Altman and Google DeepMind co-founder Demis Hassabis subsequently expressed agreement with the proposal, the complaint says. The plaintiffs characterize those statements as evidence that the companies reached a horizontal agreement to restrain competition in the development of AI products.
The complaint also points to a working group that representatives of Anthropic, OpenAI and Google allegedly had been using since July to develop an industry standards body. The filing treats that activity, together with the executives’ public statements, as evidence of a broader effort to coordinate development practices.
“The antitrust laws do not permit competitors to decide among themselves that competition is too dangerous,”
Andrew T. Tutt, Trial Lawyers for Justice, in the complaint
Consumers claim slower progress means less value
The plaintiffs argue that a coordinated slowdown would restrict product output and leave subscribers paying the same prices for services that improve more slowly than they would in an unrestricted contest. Their theory invokes Section 1 of the Sherman Act, which prohibits agreements that restrain trade, and Section 16 of the Clayton Act, which allows private parties to seek injunctive relief.
“Whether frontier AI should develop more slowly is a question for each company acting alone, or for Congress and the agencies,” Tutt wrote in the complaint. “It is not a question that four competitors may answer together.”
The case is Buist et al. v. Anthropic PBC et al., No. 26-cv-10693, in the U.S. District Court for the Northern District of California. The plaintiffs have also indicated that they intend to pursue class certification for other subscribers affected by the alleged conduct.
The legal question predates the lawsuit
Whether AI companies can coordinate on safety without violating antitrust law had already become a public concern before the complaint was filed. OpenAI asked members of Congress for guidance on whether an industry-wide slowdown could be lawful, according to reporting by WIRED.
OpenAI Chief Scientist Jakub Pachocki had argued that AI laboratories should consider “coordinating to slow down future development” until shared safety standards exist. Legal scholars cited by WIRED said the answer would depend on the details of any agreement, while warning that a joint decision to limit development could be treated as an output restriction.
The lawsuit now asks a court to examine that question in the context of specific public statements and alleged industry meetings. It does not establish that the companies reached or implemented an unlawful agreement; those allegations must be tested through the litigation.
Companies had not responded by publication
Anthropic, OpenAI, Google and SpaceX could not be reached for comment by the Daily Journal as of publication. The complaint was filed by Nicholas C. Rowley’s Trial Lawyers for Justice, which argues that safety decisions should not be controlled through private arrangements among competing commercial laboratories.
“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 in a statement.
The case puts a narrow legal issue in front of the court: whether competitors’ public agreement that AI development should proceed more slowly can become evidence of an antitrust conspiracy when paired with alleged private coordination. The complaint filed September 18 is the first formal test of that theory in this dispute.