The Federal Trade Commission (FTC) is nearing the point of sending civil investigative demands (CIDs) to frontier artificial intelligence companies, a senior FTC official told Semafor. The agency said it publicly disclosed the impending inquiry because of public health and safety concerns related to AI systems — an uncommon step, since the FTC normally does not announce investigations before issuing subpoena-like CIDs.
What the demands will ask
According to the senior official, the CIDs run dozens of questions and probe topics including what companies have said about the risks of their products and other issues related to potential consumer harm. The official also said career attorneys in the Bureau of Consumer Protection are drafting the CIDs. That contradicted reports that career staff first learned of the probe from news coverage rather than agency leadership.
Political timing and context
The report of the FTC inquiry appeared in a New York Post story at a moment when the White House sought to show it was taking AI seriously. Less than 24 hours before news of the FTC probe, FTC Chair Andrew Ferguson appeared with President Donald Trump and AI CEOs to announce a nonbinding agreement pledging industry self-regulation.
The timing and the public disclosure have prompted debate about whether the inquiry will be a deep, independent investigation or primarily a signal of action ahead of the midterm elections. Former FTC chair Lina Khan expressed skepticism, noting it strains credulity that CEOs who met at the White House would receive a credible subpoena the next morning from the same administration. William Kovacic, a former Republican FTC chair, called confirming an investigation in the press "unusual" but said it is not without precedent and can serve to alert the public that the agency is examining an issue.
Companies and stakeholders mentioned
The initial reporting named OpenAI and Anthropic, and suggested the probe could also target independent AI evaluator METR. OpenAI’s chief global affairs officer Chris Lehane said the company is trying to learn more about the matter. Anthropic declined to comment; METR did not respond to requests for comment.
Ferguson has been appointed to the White House AI task force and has stated he does not view the FTC as an "AI regulator." The Trump administration’s 2025 AI Action Plan called for review of all FTC investigations initiated under President Joe Biden to ensure they do not advance liability theories that would unduly burden AI innovation, and to consider modifying existing orders that could impede innovation.
Precedents and possible outcomes
The FTC has previously confirmed investigations in high‑profile cases, notably a probe into Meta’s privacy practices after the 2018 Cambridge Analytica revelations and its investigation of Equifax after the 2017 data breach. Agency officials typically avoid publicizing investigations to prevent prejudice against the inquiry and downstream market or operational effects for companies.
Even if this investigation proceeds, it may take a long time and not necessarily produce immediate legal action. FTC probes can be protracted, and currently there is no public finding of wrongdoing by the AI companies mentioned.
Conclusion
The FTC says it is preparing a significant inquiry into frontier AI systems and will issue CIDs containing multiple detailed questions about risk disclosures and consumer harms. The agency’s rare public announcement, its proximity to White House outreach to AI CEOs, and the administration’s stated preference for applying existing laws rather than new regulatory regimes have all contributed to debate over the probe’s intent and likely impact.
