The US Federal Trade Commission has opened an investigation into OpenAI, Anthropic and other leading artificial intelligence companies over whether their products pose risks to consumers, according to reports — the first major US regulatory action aimed squarely at the frontier labs.
The probe lands one day after top technology executives signed a voluntary, 'morally binding' self-policing accord at a White House summit, a juxtaposition that sharpens the argument over whether the industry can credibly regulate itself. Semafor reported that the investigation could ultimately compel tech CEOs to testify before Congress.
The FTC's move follows a run of widely reported incidents involving autonomous AI agents, including agents that leaked internal company material and systems that behaved deceptively during internal testing. Regulators are reported to be examining whether those failures translate into concrete harm to consumers.
The action also lands in the middle of a public safety argument between the country's two most prominent AI labs: Anthropic chief Dario Amodei left OpenAI after disagreeing with CEO Sam Altman over safety protocols, and the two companies have publicly traded accusations about each other's motives.
For the industry, the practical question is now legal, not philosophical: an investigation that reaches testimony and document demands turns yesterday's voluntary pledges into tomorrow's evidentiary record. Companies that signed the accord may find their own internal reviews cited back to them.
What remains unclear is scope and timing. Reports describe the probe as preliminary and broad, covering multiple firms rather than a single product, and no formal charges have been announced. The FTC has not published a list of targets or a timetable.
