FTC proposes policy targeting AI companies that suppress warnings about inaccurate outputs

The Federal Trade Commission has proposed a policy statement aimed at AI companies that market systems while allegedly suppressing accurate information about those systems’ accuracy or limitations.

The agency posted the proposed statement on June 30, 2026. It applies the FTC Act’s Section 5 ban on deceptive conduct to the treatment and marketing of information about artificial intelligence systems, according to the commission’s document.

The proposal does not itself impose a penalty, identify a company as liable or create a binding AI regulation. Instead, it sets out an enforcement theory the agency could use when evaluating conduct involving claims about AI performance and limitations.

What the proposal addresses

The policy statement concerns companies marketing AI systems that allegedly manipulate systems or marketing practices to suppress accurate information about the systems’ accuracy or limits. The FTC’s focus is therefore not simply whether an AI system produces an inaccurate output. The approved agency materials describe a concern about whether companies obscure or suppress information that would help users understand those limitations.

That distinction matters. The proposal does not state that every inaccurate AI output violates Section 5, and it is not an adjudicated finding that a particular company engaged in deceptive conduct. No named company is identified as having violated the proposed policy.

The FTC’s stated theory could affect how generative AI companies and other developers design systems, describe their products and handle information about safety and accuracy. It may also shape compliance practices for companies operating in U.S. commerce, although the proposal itself would not immediately require a particular product change.

Public comment deadline

The commission sought public comments through July 31, 2026. The deadline gave companies, users, researchers and other members of the public an opportunity to respond to the proposed policy statement before the agency determines what action, if any, should follow.

The source packet does not establish whether the FTC finalized or withdrew the proposal by Aug. 6, 2026. The comment deadline should therefore not be treated as proof that the policy became final. The document remains described as a proposed policy statement rather than a final rule.

Part of a broader FTC technology docket

The FTC’s technology docket also lists ongoing matters involving artificial intelligence, privacy, advertising and competition. Those matters are separate from the proposed accuracy policy and should not be treated as a single enforcement action.

A separate Justice Department antitrust matter shows another way federal agencies are examining algorithmic practices. In a July 6, 2026 announcement, the department said it had reached a proposed settlement with Willow Bridge, described as one of the country’s largest landlords. The settlement would prohibit algorithmic coordination and the exchange of competitively sensitive data in rental markets.

That Justice Department matter does not establish a violation under the FTC’s proposed AI-accuracy policy. It provides contemporaneous federal context for scrutiny of how companies use algorithms, data and automated systems in commercial markets.

What happens next

The next known step for the FTC proposal was the close of the public-comment period on July 31. The approved materials do not provide a later final decision, enforcement action or effective date.

For AI developers and users, the immediate significance is the signal about how the FTC may view the suppression of information concerning AI accuracy and limitations. Whether that signal becomes a final agency policy or contributes to a future enforcement action remains unresolved in the available record.