4 AI Giants Face Sherman Act Section 1 Suit Over Output Restraint

For antitrust and regulatory counsel, the most important question is not whether AI should be paced for safety, but whether parallel safety statements and working-group meetings cross the line from lawful information exchange into a Section 1 output restraint. The complaint frames the July 2026 statement and Dario Amodei's September 12 essay as evidence of an ongoing conspiracy, while Amodei openly acknowledged the antitrust risk and sought an exemption—a factual pattern likely to shape class certification and rule-of-reason analysis.

Four of the most prominent frontier AI developers — Anthropic, OpenAI, SpaceXAI, and Google — are now defending a proposed class action in the United States District Court for the Northern District of California that accuses them of violating Section 1 of the Sherman Act by agreeing to slow the pace of AI model development. The four named plaintiffs subscribe to Claude, ChatGPT, Grok, or Gemini and seek to represent a nationwide class of paid users. Their central allegation is not about safety failures or product defects, but about a coordinated restriction on output: if rival labs jointly decide that their systems should improve more slowly than each would choose on its own, paying customers receive less capable products than unfettered competition would deliver. The complaint argues that such a collective restraint falls squarely within the type of agreement antitrust law has long prohibited, even when the underlying technology is novel.

“

OpenAI CEO Sam Altman and Google DeepMind co-founder Demis Hassabis each signaled support; Altman said OpenAI would take an early step, while Hassabis called the essay the right path forward.

The alleged timeline is central to the case. According to the complaint, the public phase unfolded on September 12, 2026, when Dario Amodei published an essay arguing that frontier labs should jointly set limits on unchecked AI progress so that no single firm would lose commercial ground by acting alone. OpenAI CEO Sam Altman and Google DeepMind co-founder Demis Hassabis each signaled support; Altman said OpenAI would take an early step, while Hassabis called the essay the right path forward. But the suit contends the alignment began months earlier. It points to a July 2026 statement signed by senior figures at several frontier labs that acknowledged intense competitive pressure against any company that slowed by itself and urged governments to support a broader effort to moderate automated development. The complaint treats those earlier contacts, plus subsequent working-group meetings among Anthropic, OpenAI, and Google, as the start of the coordination.

Legally, the case will likely turn on whether the plaintiffs can show an agreement to restrain output rather than merely parallel safety commitments. Independent decisions to slow development are lawful, but an actual horizontal agreement among competitors to restrict output is a classic antitrust violation. The plaintiffs argue the July 2026 statement and the September 12 public endorsements supply evidence of a continuing conspiracy. Notably, Amodei acknowledged the antitrust risk in his essay and indicated he hoped the government would make an exception, which may suggest awareness that the coordination could be problematic. The Trump administration, however, rejected the idea, with the president calling warnings of AI destroying humanity a hoax, removing the political backing that might have supported an antitrust exemption.

The market implications are significant. The complaint says the alleged agreement would reduce the value consumers get for paid AI subscriptions. If the case survives a motion to dismiss, discovery could expose internal communications among the labs about pacing decisions. That could chill voluntary safety coordination in the industry, because frontier developers may become reluctant to discuss model development limits outside a formal regulatory framework. In the near term, subscribers of Claude, ChatGPT, Grok, and Gemini are left asking whether they are receiving slower progress than they paid for. The case also underscores a broader fault line: without a federal framework, private safety coordination by competitors may be illegal, but unilateral slowdowns may be commercially untenable in a highly competitive market.