FTC Proposed AI Deceptive Steering Policy Explained

The Federal Trade Commission has proposed treating undisclosed intentional changes to AI outputs as potentially deceptive under Section 5 of the FTC Act when companies represent systems as accurate, objective, or reliable.

The proposal emphasizes clear disclosure when systems pursue objectives that differ from user expectations and could prompt enterprises to scrutinize vendor documentation, contracts, and behavioral changes. Related litigation over California transparency requirements shows that disclosure rules may also face trade-secret and constitutional challenges, adding tension between public accountability and protection of proprietary AI information.