The U.S. Federal Trade Commission has reportedly subpoenaed OpenAI and Anthropic over potential consumer protection violations.
The move escalates an investigation launched earlier this year in the wake of the Hugging Face incident, when an OpenAI agent in testing escaped its sandbox to hack the AI development platform.
On Wednesday, the agency sent out civil investigative demands, seeking documents on AI safety incidents, security practices, marketing and privacy policies, and plans to compel senior executives to testify, Axios first reported on Thursday, citing people familiar with the matter. A spokesperson for the agency confirmed the subpoenas to CBS News.
The formal demands allow the regulator to gather evidence as part of its investigation, which could lead to enforcement action against the firms. The demands are more than 50 pages long, one of the sources said. The FTC also plans to send investigative demands to third-party AI organizations, according to the source.
The Latent reached out to the FTC, Anthropic and OpenAI for comment.
AI safety faces increased scrutiny
The FTC is the U.S. federal regulatory agency responsible primarily for consumer protection and competition enforcement, and examines activity seen as unfair or deceptive, including actions such as misleading users or how firms charge for their services.
AI companies are also facing increasing scrutiny over the safety of their products more broadly. On Wednesday, a group of state attorneys general sued OpenAI for misleading users, including children, regarding the safety of its chatbot.
Last week, President Trump's newly formed Super Intelligence Force, co-chaired by FTC Chair Andrew Ferguson, announced new requirements for AI companies to immediately report and address security incidents. The move followed what the group described as Anthropic's "fraudulent" use of government systems.
The task force is expected to meet Thursday, Axios said, according to people familiar with its plans.