FTC seeks comments on “suppression of AI accuracy” under Section 5

The U.S. Federal Trade Commission (FTC) is asking the public to weigh in on a proposed policy statement aimed at a practical AI problem: what happens when a company markets an AI system as accurate, objective, or reliable, but the system’s output is limited, altered, or steered in ways consumers may not reasonably expect.

The proposal was published in the Federal Register on July 7, 2026. The FTC says the public has until July 31, 2026 to submit comments.

The core issue in plain English

In the FTC’s framing, the risk is a mismatch between what consumers are led to believe about an AI product and what the system actually does behind the scenes. For example, a company may represent the tool as providing accurate results for a task, but configure it so that certain outputs are suppressed, revised, or guided toward undisclosed objectives.

Why this falls under FTC Act Section 5

The FTC is using its Section 5 authority over “unfair or deceptive” acts or practices as the legal hook. The FTC’s question for this policy statement process is essentially whether undisclosed output steering or suppression could make accuracy-related marketing or representations deceptive under the standards consumers would reasonably rely on.

What companies should be checking now

• Accuracy and reliability claims: Are marketing statements about “accuracy,” “truthful outputs,” or similar reliability promises consistent with how the system behaves?
• Disclosures about limits: If the system suppresses or steers outputs in certain situations, are those boundaries communicated clearly enough for users to understand what they’re getting?
• Testing and validation: Does the company evaluate performance in a way that reflects what the user will actually see—rather than only what the model could produce under ideal conditions?

What readers should watch next

Right now, the main deadline is July 31, 2026 for comments. After the FTC reviews submissions, the key question will be whether the agency issues a final policy statement that further clarifies how Section 5 deception analysis applies to “suppression of accuracy” scenarios.

Bottom line: this is not a finding that a specific company violated the law—it’s a proposed framework the FTC is considering, based on input from businesses, consumers, and other stakeholders.

Sources

• FTC press release (July 2026): FTC seeks public comment on policy statement addressing AI accuracy
• Federal Register (GovInfo): July 7, 2026 issue (AI accuracy policy statement notice)