What happened
On July 1, 2026, the Federal Trade Commission published its 'Proposed Policy Statement Concerning the Suppression of Accuracy in Artificial Intelligence Systems' and opened a public comment period. The statement (dated July 1, 2026, published on ftc.gov) asserts that AI companies manipulating the behaviour of their AI systems contrary to consumers' reasonable expectations for objectivity and accuracy may violate Section 5 of the FTC Act as deceptive practices. The statement explicitly warns that 'steering' an AI system's outputs to advance undisclosed ideological objectives — including attempts to comply with state AI bias laws such as Colorado's revised Artificial Intelligence Act — could constitute unlawful consumer deception even if the steering is done in a good-faith effort to comply with state law. It was issued pursuant to direction from Executive Order 14365 (Dec 2025), which directed FTC to clarify the application of federal consumer protection law to AI accuracy. The FTC is now seeking public comment before finalising.
Why it matters
This is the most significant FTC AI enforcement signal since the change of administration. If finalised as proposed, it would: (a) expose AI companies to Section 5 liability for altering model outputs to reduce bias or reflect diversity if not disclosed; (b) create direct tension with state AI laws (Colorado, California) that require bias mitigation; and (c) establish the FTC as the federal enforcer against 'ideological manipulation' of AI outputs. The statement is explicitly framed as part of EO 14365's national preemption strategy against state AI laws. AI deployers offering consumer-facing products face acute compliance risk if they tune models for bias/fairness without clear disclosure.
Action needed
Submit public comments before the comment deadline (check FTC docket for date). Immediately audit AI product disclosures: any model behaviour steering for bias, safety, or ideological balance that is not clearly and prominently disclosed to users creates Section 5 deception exposure. Legal teams should assess tension between this proposed statement and compliance obligations under Colorado SB 205/SB 189, California AB 2013, and similar state AI laws.