What happened
On Wednesday 15 October, the US Federal Trade Commission sent civil investigative demands to OpenAI and Anthropic, Axios reported, citing a person familiar with the matter. Civil investigative demands work like subpoenas: the FTC uses them to gather evidence before deciding whether to sue.
According to Axios, each demand runs to more than 50 pages. They ask for documents on security incidents and safeguarding practices, as well as marketing materials and privacy policies. The FTC reportedly plans to send demands to other AI organisations later and to compel senior AI executives to testify.
Background
The probe uses the FTC's power to act against unfair or deceptive practices. That can cover how a company describes the safety of its product or how it charges customers. Axios placed the move within a broader, tougher stance from the US administration following several reported incidents in which unreleased models behaved aggressively or deceptively during testing. The report said the President's newly created "Super Intelligence Force", which FTC Chair Andrew Ferguson co-chairs, had the week before announced an industry-wide requirement to report and fix such incidents immediately.
An OpenAI spokesperson told Axios that safety "remains a top priority" and that the company would "engage constructively with the FTC". The same day, a group of state attorneys general sued OpenAI under state consumer-protection laws (see our separate report).
Why it matters financially
An FTC investigation can end in consent orders that last for years, restrict how products are marketed or released, and bring civil penalties if they are later breached. For the two labs, and for the wider sector once third parties receive demands, the immediate costs are legal work and document production. The bigger risk is that federal conditions could be imposed on how frontier models are tested and launched. No penalty figures have been reported.