A Look at the FTC's Current Approach to AI Regulation

President Donald Trump signed Executive Order (EO) 14179, "Removing Barriers to American Leadership in Artificial Intelligence," on January 23, 2025, marking a significant shift in federal artificial intelligence (AI) policy. The EO declared "it is the policy of the United States to sustain and enhance America's global AI dominance in order to promote human flourishing, economic competitiveness, and national security." Id. § 2. It also directed the development of an AI Action Plan within 180 days and required federal agencies to review, suspend, rescind or revise actions taken pursuant to the now-revoked EO 14110 that conflict with the new policy framework. Id. §§ 4-5.

This Holland & Knight alert analyzes the Federal Trade Commission's (FTC) current approach to AI regulation considering that policy shift.

Post-EO 14179 Framework

The EO reset federal priorities by emphasizing U.S. leadership, rescinding policies viewed as barriers to innovation and directing the development of America's AI Action Plan. Released in July 2025, the plan framed AI as a "race for global AI dominance" based on three pillars: 1) accelerating AI innovation, 2) building American AI infrastructure and 3) leading in international AI diplomacy and security. Critically for FTC enforcement, the AI Action Plan specifically directs the Commission to:

Review all Federal Trade Commission (FTC) investigations commenced under the previous administration to ensure that they do not advance theories of liability that unduly burden AI innovation. Furthermore, review all FTC final orders, consent decrees, and injunctions, and where appropriate, seek to modify or set-aside any that unduly burden AI innovation.

Id. at 3.

This directive makes clear that AI-related enforcement actions should not impose undue burdens on innovation and existing FTC actions should be reassessed considering that objective.

This post-EO 14179 framework has produced a "dual approach" to AI enforcement. On the one hand, the FTC has curtailed "enforcement related to the actual capabilities of AI products." On the other hand, the FTC continues to pursue enforcement where companies allegedly deceive consumers through false statements related to the capabilities of their AI products. This distinction aligns with the AI Action Plan's goal of limiting enforcement theories that may impede AI innovation while preserving the FTC's authority to police deceptive or misleading commercial practices.

Proposed Policy Statement Concerning the Suppression of Accuracy in AI Systems

On July 1, 2026, the FTC's Proposed Policy Statement Concerning the Suppression of Accuracy in Artificial Intelligence Systems (Proposed Statement) addressed the Commission's concern that AI companies may steer AI system outputs in ways that are contrary to reasonable consumer expectations for objectivity and accuracy. The Commission explained that the FTC Act prohibits businesses from engaging in unfair or deceptive conduct and that AI companies may deceive consumers in violation of Section 5 of the FTC Act when they distort AI system outputs to achieve undisclosed objectives. The FTC also explained that AI companies have marketed their systems as tools designed to produce the best output possible within technological and resource constraints. Accordingly, the consumers reasonably expect AI systems to aim for truthful and accurate outputs. The proposed statement therefore frames the FTC's current AI focus as a deception issue, emphasizing whether companies mislead consumers about what their AI systems are designed to do.

The FTC draws an important distinction by recognizing that though AI can generate incorrect outputs, those outputs by themselves do not necessarily raise issues under Section 5 of the FTC Act. However, a company may still deceive consumers if it misrepresents the likelihood of inaccurate outputs or if inaccurate outputs result from a design decision to prioritize objectives contrary to users' reasonable expectations.

The proposed statement also explains how companies may reduce deception risk through disclosure. An AI company may shape consumer expectations by truthfully and clearly disclosing that its systems are designed to prioritize different objectives from what users request or otherwise expect. However, the FTC warns that such a disclosure must be clear, conspicuous and sufficient to change consumer expectations; it cannot be buried in terms of service or hidden in fine print. Thus, the FTC's guidance centers on transparency. Together, America's AI Action Plan and Proposed Statement show the FTC's current position. The FTC has moved away from broad theories that potentially treat AI tools themselves as problematic and instead focuses on whether companies have engaged in deception or other conduct that constitutes a cognizable Section 5 violation. The FTC's recent enforcement actions reflect this shift.

The FTC's Current Agenda as Shown Through Its Enforcement Actions

Recent FTC enforcement actions reflect the framework described above. The FTC continues to pursue AI-related conduct when companies make concrete misrepresentations through false statements related to the capabilities of their AI products. At the same time, the FTC has signaled that it will not impose broad restrictions on AI tools merely because they could potentially be misused. The following matters illustrate the FTC's current approach.

Rejecting Overbroad AI-Tool Liability

In In re Matter of Rytr LLC, No. C-4806, 2025 WL 4740225, at *1-2, (F.T.C. Dec. 22, 2025), the FTC addressed whether a generative AI writing service could be restricted on the theory that users might misuse it to generate deceptive consumer reviews. Rytr sells a generative AI writing service that allowed subscribers to generate written content, including consumer reviews. The prior FTC alleged that Rytr violated Section 5 of the FTC Act in two ways: 1) by providing users with the "means and instrumentalities" to generate deceptive consumer reviews and 2) by operating an unfair review-writing service that could generate numerous reviews without regard for accuracy. To resolve these allegations, Rytr consented to an order barring it from offering any service dedicated to, advertised as, promoted as or offered as generating consumer or customer reviews or testimonials.

Following EO 14179 and the release of America's AI Action Plan, the FTC reopened the matter and set aside the order. The FTC concluded that 1) the complaint did not support a cognizable Section 5 violation, 2) the order therefore did not provide any benefit to the consumers or the public, and 3) maintaining an order untethered to an actual legal violation would unduly burden AI innovation. The FTC explained that the complaint did not allege Rytr itself created deceptive marketing materials, Rytr's product was inherently deceptive, or Rytr knew or had reason to know that users would use the tool to violate Section 5. In determining that the previous order unduly burdened innovation in the AI industry, the FTC relied on now-Chair Andrew Ferguson's dissent explaining that "[t]reating as categorically illegal a generative AI tool merely because of the possibility that someone might use it for fraud is inconsistent with our precedents and common sense. And it threatens to turn honest innovators into lawbreakers and risks strangling a potentially revolutionary technology in its cradle."

Rytr demonstrates the FTC's post-EO 14179 approach. In setting aside the prior order, the FTC expressly rejected a theory of liability that treated a generative AI tool as unlawful because it could be used to facilitate deception. The FTC requires a concrete connection between the challenged AI-related conduct and cognizable Section 5 violation before imposing restrictions on AI products.

Misrepresenting an "AI-Powered" Advertising Service

In the FTC's actions against an American media conglomerate, as well as MindSift LLC and 1010 Digital Works LLC, the FTC addressed the companies' marketing of an allegedly AI-powered advertising service known as "Active Listening." The FTC alleged that the companies falsely claimed the service could "target localized ads based on conversations captured from consumers' smart devices and that consumers had opted into such targeting." According to the FTC, however, the service did not listen to consumers' conversations or use voice data at all; instead, it consisted of reselling email lists obtained from data brokers at a significant markup. The FTC further alleged the companies falsely represented that consumers had consented to the service. To resolve the allegations, the proposed orders require the media conglomerate to pay $880,000 and MindSift and 1010 Digital Works to pay $25,000 each, as well as prohibit misrepresentations about advertising service features, voice data collection and consent, and geographic targeting capabilities.

The allegations concerning the purported "Active Listening" capabilities illustrate the FTC's post-EO 14179 enforcement approach. Rather than restricting AI technologies based on speculative concerns about potential misuse, the FTC focused on specific allegedly deceptive representations regarding the service's capabilities and consumer consent.

Misleading Income Claims Ties to AI-Powered Software

In the FBA Machine/Passive Scaling matter, the FTC challenged an alleged business opportunity scheme in which FBA Machine and its owner, Bratislav Rozenfeld, allegedly falsely guaranteed that consumers could earn substantial income by operating online storefronts using AI-powered software. According to the FTC, the defendants did not deliver on those promised earnings and defrauded consumers of more than $15 million. To resolve the matter, the proposed settlement permanently bans FBA Machine and Rozenfeld from selling business opportunities, prohibits them from making the specific alleged misrepresentations or other material misrepresentations in selling products or services, and imposes a $15.7 million monetary judgment, which was partially suspended based on the defendants' ability to pay.

Rather than treating the use of AI as inherently problematic, the FTC focused on allegedly false earnings claims and representations concerning the capabilities of AI-powered software. The FTC's theory of liability rested on whether the consumers were misled about the financial results the technology could achieve, not on the underlying technology.

Misleading Claims About AI Accuracy or Efficacy

The FTC gave final approval to an order against Workado LLC after alleging that the company misrepresented the accuracy and effectiveness of its AI Content Detector, a product marketed to consumers seeking to determine whether written content was generated by AI or authored by a human. Workado allegedly represented that the tool had been developed using a broad range of materials, including blog posts and Wikipedia entries, making it more accurate for average users. According to the FTC, however, the model powering the tool was trained or fine-tuned primarily to classify academic content and therefore did not support the broader performance claims made to consumers. The final order requires Workado to refrain from making any representations that are false, unsupported or misleading.

The allegations against Workado illustrate the FTC's post-EO 14179 enforcement approach. Rather than treating AI-content detection as inherently problematic, the FTC focused on whether the company possessed adequate substantiation for its claims regarding the tool's accuracy and effectiveness. Thus, Workado reinforced the FTC's broader emphasis on combating deceptive statements relating to the capabilities of AI products rather than restricting AI innovation itself.

Misleading Claims About Conversational AI Capabilities

The FTC brought an enforcement action against Air AI, five related companies and their owners based on allegations that they misled entrepreneurs and small businesses through deceptive claims about business growth, earnings potential and refund guarantees. According to the FTC, the defendants falsely represented that purchasers were likely to earn substantial income, misrepresented the availability of refund and buy back guarantees, and made unsupported claims regarding the performance, profitability and overall characteristics of their AI-related business opportunities. To resolve the allegations, the proposed order prohibits Air AI and its operators from marketing or selling business opportunities, making false or misleading claims in connection with telemarketing or the sale of goods and services, and making earnings claims without adequate substantiation or required disclosures. It also imposes an $18 million monetary judgment.

The allegations and proposed order against Air AI further illustrate the FTC's current approach to AI enforcement. The FTC focused on allegedly unsupported representations concerning earnings potential, business growth and product performance. Thus, companies may not rely on AI-related products or services as a basis for making unsubstantiated promises. Similar to the other recent AI-related actions, Air AI demonstrates that the FTC's enforcement efforts are not directed at the underlying AI technology itself.

Conclusion

The FTC's post-EO 14179 approach to AI enforcement reflects a narrower – but not passive – regulatory posture. EO 14179 and America's AI Action Plan direct the FTC away from theories of liability that unduly burden AI innovation, and the Commission's decision to set aside the order in Rytr demonstrates a clear departure from efforts to impose liability based solely on the possibility that an AI tool could be misused. Instead, the current FTC has emphasized that restrictions on AI products should be tied to cognizable Section 5 violations and supported by concrete allegations of unlawful conduct.

At the same time, the FTC has not retreated from AI enforcement. The Proposed Statement and recent enforcement actions show continued scrutiny of companies that allegedly misrepresent the capabilities of AI-related products and services. Taken together, the AI Action Plan, Proposed Statement and FTC's recent enforcement actions suggest that the Commission's current AI agenda is best understood as a dual approach: reduced enforcement directed at AI innovation itself, coupled with continued enforcement against deception, false advertising, unsupported performance claims and other traditional Section 5 violations.

Former Holland & Knight Summer Associate Daniela Garcia Baerga also contributed to this alert.

Information contained in this alert is for the general education and knowledge of our readers. It is not designed to be, and should not be used as, the sole source of information when analyzing and resolving a legal problem, and it should not be substituted for legal advice, which relies on a specific factual analysis. Moreover, the laws of each jurisdiction are different and are constantly changing. This information is not intended to create, and receipt of it does not constitute, an attorney-client relationship. If you have specific questions regarding a particular fact situation, we urge you to consult the authors of this publication, your Holland & Knight representative or other competent legal counsel.