Four paying subscribers have sued the biggest names in artificial intelligence. The federal antitrust lawsuit landed Friday in Northern California.
It names Anthropic, OpenAI, SpaceXAI, and Google as defendants. The case could become a landmark test for AI competition. The four companies dominate the market for consumer AI assistants.
The complaint is titled Buist v. Anthropic PBC. The Associated Press first reported the filing on Friday. Each named plaintiff pays for ChatGPT, Claude, Grok, or Gemini. Lawyers filed it for a proposed nationwide subscriber class.
The AI Slowdown Agreement at the Heart of the Case
The complaint centers on Anthropic CEO Dario Amodei’s September 12 essay. Amodei urged industry-wide coordination to ‘pace the frontier’ safely.
He wrote that labs must slow AI capability improvement. The essay cited risks from increasingly powerful AI systems.
Within hours, three rival chief executives publicly endorsed the proposal. Elon Musk replied ‘Dario is right,’ according to the complaint.
Sam Altman also publicly agreed with Amodei that day. Demis Hassabis called the proposal ‘the right path forward.’
The plaintiffs say this public concurrence was actually illegal. They call it an unlawful agreement among direct competitors. The filing calls the swift endorsements evidence of collusion. The same-day responses anchor the AI slowdown collusion claim.
Why Subscribers Say the AI Slowdown Pact Hurts Them
The plaintiffs argue the deal harms consumers directly, according to CBS News. They pay the same price for slower-improving products. Slower updates mean less value for identical subscription fees. ChatGPT Plus alone costs about twenty dollars each month.
The complaint defines a market for paid frontier AI assistants. The four defendants allegedly control at least eighty percent. That consumer market already feels pressure as enterprise AI spending cools.
The complaint says antitrust law forbids private decisions limiting competition. Lead attorney Nick Rowley wants safety rules set publicly. He says private deals between dominant firms are dangerous.
Rowley warned AI could ‘quickly spin out of human control.’ The government, he argues, should set the safety guardrails.
What the Lawsuit Seeks and What Comes Next
The plaintiffs seek certification of a nationwide subscriber class. They also demand treble damages under the Clayton Act. An injunction would bar coordinated limits on release timing. It would also cover capability checkpoints and compute decisions.
Amodei anticipated antitrust objections inside his original essay. He suggested Washington grant a narrow waiver for safety talks. Amodei said officials need not join the discussions. Republican Senator Josh Hawley quickly dismissed that idea.
OpenAI already faces a mounting pile of courtroom fights. This case adds antitrust exposure to that growing list. Its rivals still race to dominate consumer AI.
None of the four companies have commented publicly yet. The case now sits before a California federal judge. Its outcome could reshape how rivals discuss safety together.
A ruling may redefine how the industry handles safety coordination. For now, subscribers watch a rare legal showdown unfold.
