FTC seeks comment on its proposed AI accuracy policy statement

The Federal Trade Commission (FTC) is seeking public comment on a proposed policy statement it says seeks to address concerns about AI companies allegedly manipulating the behavior of their AI systems “contrary to reasonable consumer expectations for objectivity and accuracy,” according to a press release.

“As the proposed policy statement explains, the FTC Act prohibits businesses from engaging in ‘unfair or deceptive’ conduct,” the release states. “The proposed statement goes on to describe how AI companies that distort their systems’ outputs to achieve undisclosed ideological objectives could be deceiving consumers in violation [of] Section 5 of the FTC Act. Such conduct, it explains, may be at odds with explicit and implicit representations made to consumers about the effectiveness and suitability of AI systems for various tasks.”

The release notes that the FTC Commission vote authorizing the Federal Register notice was 2-0. Five commissioners, nominated by the president and confirmed by the Senate, lead the FTC, according to the commission’s website. Only two commissioners are listed.

Chairman Andrew N. Ferguson said that the FTC wants to hear from businesses and consumers on “the subversion of AI systems for ideological ends.”

“This crucial input will help the commission formulate a final policy that advances President Donald Trump’s goal of expanding America’s global dominance in artificial intelligence,” he said in the release.

FTC notes that some state laws seek to impose their own patchwork of regulations on AI. According to the commission’s provided example, Colorado’s Artificial Intelligence Act appears to coerce companies into altering the output of their AI models to comply with and advance the state’s ideological objectives.

“…AI companies might steer their systems’ outputs toward objectives other than those that consumers ask for or expect,” the proposal states. “A company could be tempted, for example, to abuse consumer trust by training a model surreptitiously to produce ideologically motivated distortions in a response to a factual question, such as to correct what the developer believes are ‘historical injustices’ in the facts. Or a company could be pressured by a state law to alter its technology’s outputs.

“It is predictable that an AI company might suppress accuracy and interpose other objectives, such as so-called ‘equity,’ to avoid liability under this [Colorado] law, but fail to disclose these ulterior objectives in order to hide the loss of accuracy they necessitate. AI companies may also be subject to public pressure to surreptitiously avoid politically inflammatory outputs, as well as the temptation to clandestinely modify their AI systems to further their own or their employees’ political agendas. In light of the above, the Commission believes that AI companies that steer the outputs of their AI systems toward unexpected objectives, and away from the objectives set by or reasonably expected by users, are likely to deceive.”

The FTC press release adds that the commission’s proposed policy statement explains such a law is “‘impliedly preempted to the extent it conflicts with a federal regulatory scheme.'”

In a December executive order, Trump directed the FTC to issue a policy statement addressing the legal implications of state laws that require alteration of the “truthful outputs of AI models.”

The proposed policy statement will be published in the Federal Register. The public will have until July 31 to submit comments on it. Once processed, the FTC says the comments will be posted on Regulations.gov.

In September, the FTC launched an investigation into consumer-facing AI-powered chatbots, stating that the purpose was to safeguard children online.

The FTC issued orders to seven companies that use AI-powered chatbots to determine whether potential negative impacts of their use by children and teens are sufficiently measured, tested, and monitored.

“AI chatbots can effectively mimic human characteristics, emotions, and intentions, and generally are designed to communicate like a friend or confidant, which may prompt some users, especially children and teens, to trust and form relationships with chatbots,” the FTC said in a Sept. 11, 2025 press release.

“The FTC inquiry seeks to understand what steps, if any, companies have taken to evaluate the safety of their chatbots when acting as companions, to limit the products’ use by and potential negative effects on children and teens, and to apprise users and parents of the risks associated with the products,” the release states.

Collision Hub CEO Kristen Felder posted on LinkedIn, encouraging collision repair facilities to comment on how AI can negatively affect vehicle damage estimates.

“The current policy statement is primarily focused on the suppression or manipulation of accurate AI outputs for undisclosed objectives (reducing claims severity),” Felder wrote. “That means shops should provide specific, documented examples and clearly explain how the AI system’s output appeared inaccurate, distorted, incomplete or inconsistent with what consumers were told about the system.”

She added that commenters should be more specific and provided an example:

“Do not simply say, ‘The estimate was low.’ Explain:

• • “What the AI estimate omitted
  • “What the shop later discovered
  • “What documentation was required to correct it
  • “Whether the omission affected safety
  • “How it affected the consumer or your business
  • “Whether the same pattern has occurred repeatedly

“Photos, estimates, supplements, OEM procedures, scan results and other documentation may help show the difference between the AI output and the vehicle’s actual repair requirements,” Felder wrote.

As of Friday afternoon, 105 comments had been submitted. Those who wish to comment can do so here.

The National Association of Insurance Commissioners (NAIC) rolled out an AI Evaluation Tool pilot program earlier this year to help regulators understand how insurers use AI in 12 states.

The program started in March and will run to September.

“It provides a structured way for states to review AI systems, promote transparency, and identify where additional oversight, training, or improvements may be needed,” an NAIC document states.

States included in the pilot are: California, Iowa, Rhode Island, Colorado, Louisiana, Vermont, Connecticut, Florida, Maryland, Pennsylvania, Virginia, and Wisconsin.

Images

Featured image credit: Ward DeWitt/iStock