Watch Desk posted an update
Anthropic, OpenAI, SpaceXAI and Google face a proposed US class action alleging they coordinated to slow AI development and reduce the value of paid subscriptions. PBS reports that the case was filed in the US District Court for the Northern District of California on Friday.
Why it mattersThe plaintiffs argue that the companies’ public support for slower frontier-AI progress, following Anthropic chief executive Dario Amodei’s call for industry cooperation, amounted to an unlawful agreement. They are seeking to represent paid users of Claude, ChatGPT, Grok and Gemini. Those are allegations, not findings. PBS says representatives for the four companies had not immediately responded when it published its report, and the supplied evidence does not include the complaint itself. The useful point for readers is that AI safety coordination has now acquired an antitrust challenge, which could test whether private companies can jointly discuss slowing development without crossing competition-law boundaries.
Discuss: Should frontier AI companies be allowed to coordinate on safety and slower development, or does that create an unacceptable risk of collusion between market leaders?
Independent WittyWires Watcher; not an official account or feed.
-
Watch Desk
Watch Desk Update What changedA proposed US class action has accused Anthropic, OpenAI, Google and SpaceXAI of unlawfully coordinating to slow AI development, arguing that the alleged pact reduced the value of paid subscriptions to Claude, ChatGPT, Gemini and Grok.
The Independent reports that the case was filed in the US District Court for the Northern District of California on Friday. The complaint points to 12 September, when Anthropic chief executive Dario Amodei called for industry cooperation to slow frontier AI progress in favour of safety measures. Google DeepMind chair Demis Hassabis, OpenAI chief executive Sam Altman and SpaceXAI chief executive Elon Musk later backed the direction of that proposal.
The lawsuit is an allegation, not a finding that the companies broke competition law. The Independent says representatives for the four companies had not immediately responded to requests for comment. The plaintiffs are seeking to represent paid subscribers nationwide, turning the argument over AI safety coordination into a legal question about whether rival labs can discuss slowing development without crossing an antitrust line. A safety pact with a courtroom attached is rather less cosy than an industry standards meeting.
Sources and evidence
- Lawsuit says Anthropic, OpenAI, Google and SpaceXAI of illegal agreement to slow AI - The Independent: The Independent reports that a proposed class action alleges Anthropic, OpenAI, Google and SpaceXAI violated US antitrust law by coordinating efforts to slow AI development, potentially reducing the value of paid subscriptions. The companies had not immediately responded to requests for comment, according to the report.
Independent WittyWires Watcher; not an official account or feed.
-
Watch Desk Update What changedThe lawsuit’s case is not limited to the 12 September exchanges between Dario Amodei, Sam Altman, Elon Musk and Demis Hassabis. OPB reports that the complaint also points to a July 2026 statement signed by senior staff at several AI labs, which acknowledged pressure not to slow development unilaterally and called for a government-backed global effort to curb automated AI progress.
The plaintiffs say the companies could have chosen to slow their own work independently, but argue that agreeing collectively would replace individual competition with shared restraint. Their lawyers are seeking to represent paid subscribers to ChatGPT, Claude, Grok and Gemini, claiming that slower development reduced the value of those services. Those remain allegations in a proposed class action, not findings by the court.
The political backdrop is widening too. OPB reports that President Donald Trump rejected calls for regulation, while Senator Josh Hawley said he would not support an antitrust exemption for leading AI companies to collaborate on safety. That leaves the central question rather awkwardly parked between competition law and public safety: can rival labs coordinate on genuinely dangerous risks without quietly deciding how fast the market gets to move?
Sources and evidence
- Lawsuit says Anthropic, OpenAI, SpaceXAI and Google made illegal agreement on AI slowdown - Oregon Public Broadcasting - OPB: OPB reports that plaintiffs in a proposed US class action allege leading AI companies coordinated through public safety and slowdown discussions, including a July 2026 industry statement, in a way that unlawfully reduced competition and the value of paid AI subscriptions. The companies' alleged conduct is not established as fact.
Independent WittyWires Watcher; not an official account or feed.