The FTC’s AI Moment

A summary of last week’s FTC nominations hearing

By Andy Jung, Associate Counsel at TechFreedom.

Halfway through the nomination hearing for newly appointed FTC commissioners Melissa Holyoak and Andrew Ferguson, Senator Ed Markey exclaimed: “We’re having an AI moment!” Indeed we are. Artificial intelligence came up several times during the hearing last Wednesday.

Committee Chair Maria Cantwell (D-WA) kicked off by mentioning AI in her opening statement: “The FTC is investigating fraudulent practices resulting from artificial intelligence.” The Chair probably had in mind the FTC’s recent blog post on chatbots and deepfakes. Or perhaps the Commission’s not-so-non-public investigation of OpenAI.

Ranking Member Ted Cruz (R-TX), a Former Director of the Office of Policy and Planning at the FTC, also mentioned AI in his opening statement. But Senator Cruz’s line of questioning was much more critical of the agency.

“Today’s FTC is sadly unrecognizable from the FTC I knew,” opined Cruz. He attacked the Commission for “pursu[ing] a plainly partisan agenda well outside the agency’s legal authority and mission.” Specifically, Cruz criticized the FTC for seeking “unauthorized regulation of artificial intelligence in a manner that infringes on constitutionally protected speech and will chill entrepreneurship.”

Cruz is likely referencing the FTC’s civil investigative demand to OpenAI, which requires the firm to report “unauthorized” attempts by users to “bypass filters or manipulate” its AI tools into performing prohibited actions. Forcing OpenAI to list these users would threaten their right to anonymous speech.

Later on, Senator John Thune (R-SD) asked the nominees if they “believe the FTC has the authority to regulate AI, or should it be left to Congress?” The question set the table for a revealing discussion of the limits of FTC authority over AI.

Commissioner Slaughter answered first. "Existing law does apply to new technologies,” like AI. “The FTC Act has adapted to new technologies over its hundred-plus-year history.” Commissioner Slaughter clarified: “our job is to look at the laws that we have, which primarily prohibit unfair and deceptive acts and practices and unfair methods of competition. Some of those may be triggered by some deployments or uses of AI.”

But, Slaughter added, “it may be that new authority is also needed or new laws are needed, and that is entirely up to Congress to debate and decide.” She concluded by reiterating that there may be AI issues “that Congress deems are problematic, that go beyond what the FTC Act covers, and that is up to [Congress] to decide whether additional rules are required.”

Commissioner-nominee Ferguson, a Republican, agreed with Commissioner Slaughter, a Democrat. “The FTC doesn’t regulate specific industries,” he said “It enforces rules and laws across all industries. If the conduct that AI facilitates violates those laws or rules, the FTC should act. But insofar as we’re contemplating a grand regulatory framework for AI, I think that’s squarely within the purview of Congress.”

Slaughter’s and Ferguson’s cautious comments on Congressional authority over AI reflect a growing awareness throughout Washington of the Supreme Court’s recent decision in West Virginia v. EPA. Applying the major questions doctrine, the Court curtailed the authority of the Environmental Protection Agency. The majority explained that, “in certain extraordinary cases[,]” an agency “must point to clear congressional authorization for the power it claims.” The nominees seem to recognize that far-reaching AI regulation would likely be an extraordinary undertaking requiring clear authorization.

In response to Senator Thune, Commissioner-nominee Holyoak voiced concern over “artificial intelligence in frauds and scams” like robocalls and phishing emails. Holyoak added: “I think we apply the laws that we have, but certainly if I’m confirmed we’ll talk to [Congress] about if there are gaps that we think that we need to address.”

Thune fired back, rebuking the nominees’ claims of authority over AI: “I have serious concerns about the FTC regulating or making any attempt to regulate AI.” Instead of “heavy-handed regulations,” Thune proposed “a light-touch approach that avoids unnecessary government intervention, so that innovation can continue to flourish.”

AI also came up during the hearing in the context of data privacy. In his opening statement, Senator Cruz referenced the FTC’s ongoing rulemaking on so-called ‘commercial surveillance.’ He condemned the Commission for proposing “broad data privacy and security rules without congressional authorization.”

Senator Marsha Blackburn (R-TN) added that data privacy and AI are two areas receiving congressional attention, so “we were all a little bit surprised that last November the FTC moved forward with a rulemaking on commercial surveillance and data security.” And Senator John Hickenlooper (D-CO) noted that “strong data privacy legislation would empower consumers, creators, and help us grow our modern AI-enabled economy.”

During the hearing, all three nominees asserted FTC authority to police unfair or deceptive AI practices affecting commerce. Commissioner Slaughter also hinted at FTC authority over unfair methods of competition related to AI, as did the Commission in a blog post published in June.

When policing AI, the Commission’s safest bet would be to focus on case-by-case adjudication, applying existing authority to discrete facts—as opposed to giant rulemakings, which are far more likely to veer into the purview of Congress. Ultimately, the nominees and senators all agreed that the authority to enact a grand regulatory framework for AI lies with Congress alone.