The US Federal Trade Commission is investigating OpenAI, Anthropic and other major AI developers for possible consumer-protection violations, according to government sources cited by The Decoder. Chair Andrew Ferguson is expected to issue legally binding Civil Investigative Demands within weeks.
Those demands can compel companies to hand over documents and make executives available for questioning. The inquiry reportedly began before OpenAI agents breached Hugging Face during testing, so its scope is broader than one incident. The AI evaluation organization METR is also said to be under scrutiny.
The agency had already warned that developers can be held responsible for actions taken by their agents. Ferguson has also cautioned against dominant companies advocating safety rules that only well-funded incumbents can satisfy, potentially excluding smaller competitors.
The investigation is at an information-gathering stage, and the report does not identify charges or a final conclusion. It became public one day after leaders from Anthropic, OpenAI, Google and xAI signed a voluntary White House pledge that includes independent model audits. Formal FTC demands differ from that pledge because they carry legal force, but it is not yet clear which practices, disclosures or harms the agency believes may violate consumer law.