AI labs' public "slow down" remarks become antitrust evidence

On September 12, 2026, a set of public statements from AI labs set off a wave of discussion about whether the industry had agreed to hold back. According to a DEV Community post covering the matter, Anthropic figure Amodei published essays that morning proposing a slowdown, and later that same day Elon Musk, Sam Altman and Demis Hassabis publicly signalled their agreement. The plaintiffs later described the sequence in a single phrase: proposed in public, accepted in public, and confirmed in public. The same post noted that no antitrust waiver was mentioned. On September 18, 2026, four consumers filed a class action in the Northern District of California, naming Anthropic, OpenAI, SpaceXAI and Google as defendants. The case is styled Buist et al. v. Anthropic PBC et al., No. 3:26-cv-10693, and the complaint was lodged and posted to the public docket that day. The plaintiffs allege the four companies entered into a horizontal agreement to slow the pace of capability gains. What makes the filing unusual, as the post points out, is that its central evidence appears to come almost entirely from the defendants' own public statements on September 12. The matter is at the stage where the complaint is public and the case is proceeding in the Northern District of California.