U.S. Federal Trade Commission Issues Subpoenas to OpenAI and Anthropic
nashnova research
The FTC issued 50-plus-page civil subpoenas to both OpenAI and Anthropic on Wednesday, demanding documents on safety incidents and consumer protection — marking a shift from rhetoric to formal evidence-gathering in AI regulation.
What exactly is the FTC looking for?
The subpoenas are Civil Investigative Demands — a legal tool the government uses to collect evidence before potential litigation. Each runs over 50 pages.
They require four categories of documents: safety-incident records, information-security practices, marketing materials, and privacy policies.
This means → the FTC is not broadly "monitoring" AI safety; it is pulling evidence item by item, zeroing in on whether consumer-protection laws have been violated.
Does the probe stop at documents?
No. Sources say the FTC also plans to compel executives from both companies to testify about their products.
Subpoenas for third-party AI organizations will follow — the investigation is expanding industry-wide.
In plain terms = this is not a spot check on two firms; it is the opening move of a systematic sweep.
Why now?
On the same day, multiple state attorneys general jointly sued OpenAI, alleging it misled users — including minors — about chatbot safety. Administrative and judicial pressure is rising in tandem.
FTC Chair Andrew Ferguson's "Super Intelligence Force" responded last week to allegations that Anthropic "fraudulently" used government systems, announcing a mandatory industry-wide requirement: safety incidents must be reported and remediated immediately.
This reflects a shift in regulatory logic — from "react after the fact" to "build the evidence chain first." The subpoena itself is the signal.
What does this mean for the industry?
The FTC's jurisdiction covers unfair or deceptive acts that harm consumers, including billing practices and potential user deception.
Ferguson has stated publicly that AI systems should not ship if they fail to meet safety standards — a position shared by some industry executives.
This means → the next focal point is whether OpenAI and Anthropic can mount an effective response through document review and executive testimony. The outcome will set the regulatory tone for the entire AI industry.
市场有风险，内容仅供研究参考，不构成投资建议。