The Federal Trade Commission said today that artificial intelligence companies whose conversational systems generate responses reflecting ideological objectives may be violating federal statutes. The statement was published as part of a proposed policy document laying out how the agency plans to apply its enforcement powers to the AI sector.
In the agency's account, directing chatbots away from outputs that could be viewed as discriminatory against particular groups could run afoul of Section 5 of the Federal Trade Act. That provision, the FTC noted, bars unfair or deceptive business practices.
The agency also flagged a potential legal tension between its own authority and state-level measures. The FTC said that following a Colorado law intended to curb AI-enabled discrimination in hiring and other consequential decisions could itself create a conflict with the FTC Act.
The statement further referenced ongoing political concerns about AI bias. President Donald Trump and other conservatives have accused AI chatbots of demonstrating political bias against them, and the agency framed this matter as another example of conservatives turning to federal power to pursue those complaints.
As part of the rulemaking process, the FTC will accept public comment on the proposed policy until July 31. The agency's notice invites stakeholders and members of the public to submit feedback within that window.
Context and implications
The proposed policy is intended to clarify how the FTC will use its existing legal authorities when dealing with AI developers and operators. It draws attention to two specific legal concerns: potential violations of Section 5 of the Federal Trade Act when AI systems are tuned to produce ideologically driven outputs, and possible conflicts between complying with state anti-discrimination measures and obligations under the FTC Act.
The agency is seeking input through a formal comment period that closes July 31, signaling that the proposal remains subject to change based on the responses the FTC receives.