Class action: Anthropic, OpenAI, Google, xAI colluded on an illegal 'AI slowdown' pact
Four paid ChatGPT/Claude/Gemini/Grok subscribers filed a proposed antitrust class action (Buist et al. v. Anthropic PBC et al., N.D. Cal., filed 18 Sept 2026) alleging the labs' 12 Sept public exchange on pacing AI development, triggered by Dario Amodei's essay urging industry-wide safety coordination, amounted to an illegal agreement to slow development and reduce the value of paid subscriptions. For advisory clients this is a live illustration of the antitrust exposure in any cross-lab 'safety coordination' talk: a genuine AI-safety rationale does not itself immunise coordinated conduct from Sherman Act scrutiny, and any client considering joint industry statements on capability pacing needs antitrust counsel alongside AI-safety counsel before speaking publicly.
Source: Associated Press, via Fortune
Unsealed NYT v. OpenAI/Microsoft filing: internal admissions cut against the fair-use defence
Judge Stein unsealed a New York Times brief in the NYT/Daily News/CIR copyright suit against OpenAI and Microsoft showing internal admissions that go to the heart of the fair-use market-harm test: a Microsoft director privately called the scraping practice theft of an extraordinary scale, Microsoft's own data reportedly showed Copilot cutting click-throughs to NYT content by as much as 93%, and OpenAI's head of ChatGPT is said to have described chatbot outputs internally as substitutive for the original journalism. Much of the detail comes from the Times' own brief rather than the still-sealed underlying exhibits, so treat the specifics as reported, not confirmed from the primary document. The practitioner point: courts assessing AI fair-use defences are increasingly going to weigh a provider's own internal characterisation of its practices, not just the technical training process, which is a live risk for any client's AI vendor-diligence and content-licensing decisions.
Source: TechCrunch (Rebecca Bellan), reporting on the unsealed court filing
Unsure which of these developments applies to your AI systems?
Begin in writing →This briefing is general information, not legal advice, and does not create an advisor–client relationship. Summaries are original; follow source links for the full record. Adesanya AI Advisory — Abdulwahab B. Adesanya, Barrister-at-Law (Nigeria).