The Federal Trade Commission is the United States' federal consumer-protection and competition regulator. It enforces Section 5 of the FTC Act, which prohibits unfair or deceptive acts or practices, and shares federal antitrust enforcement with the Department of Justice. The FTC has no AI-specific statutory mandate, but its general Section 5 authority, its Section 6(b) study power, and new duties assigned to it by statute (notably the TAKE IT DOWN Act) have made it one of the most active US agencies on AI in practice. Andrew Ferguson has chaired the Commission since January 2025.
Authorities relevant to AI
The FTC reaches AI through three instruments. Section 5 unfair-or-deceptive-practices authority applies to AI products as to any consumer product, and is the hook for deception, dark-pattern, and safety-representation theories. Section 6(b) of the FTC Act lets the Commission compel companies to answer detailed studies without an enforcement predicate. And Congress has begun assigning the FTC AI-adjacent enforcement duties directly: the TAKE IT DOWN Act treats violations of its notice-and-takedown duty as Section 5 violations, with no private right of action (see TAKE IT DOWN Act).
Companion-chatbot 6(b) inquiry (2025–)
On September 11, 2025 the FTC issued 6(b) orders to seven operators of consumer-facing AI chatbots — Alphabet, Character Technologies, Instagram, Meta, OpenAI, Snap, and xAI — seeking information on how they monetize engagement, measure and mitigate harms, and handle minors (Source: ftc.gov). It is the most prominent federal action on the companion-chatbot category to date (AI Companions, Companion Chatbot Harms — Cross-Cutting Analysis). In February 2026, Senators Kirsten Gillibrand and Rick Scott wrote to Ferguson urging the inquiry to also examine how companion products affect older at-risk users (Source: gillibrand.senate.gov).
TAKE IT DOWN Act enforcement (2026–)
Chair Ferguson sent a stakeholder letter on May 11, 2026 to fifteen named platforms ahead of the Act's May 19, 2026 Section 3 compliance deadline, disclosing a civil-penalty figure of $53,088 per violation and an expected cross-platform hash-sharing baseline (FTC Take It Down Act Stakeholder Letter (Ferguson, May 11, 2026)). Enforcement began on the May 19 effective date; on May 21 the agency sent warning letters to twelve unidentified websites over possible violations, its first concrete enforcement step under the Act (Source: insideaipolicy.com). Detail on the regime is at TAKE IT DOWN Act.
AI preemption policy statement (2026)
The FTC published the proposed policy statement on July 1, 2026, formally opening a public comment period and arguing that AI companies that distort outputs toward undisclosed ideological objectives could violate Section 5's prohibition on deceptive conduct (Source: ftc.gov). Issued under President Trump's December 2025 executive order on a national AI policy framework, the statement addresses whether AI companies that steer their models' outputs violate 15 U.S.C. § 45, claims preemption authority over state laws that would require altering "truthful" model outputs — characterized as "ideological" state AI laws — and singles out the Colorado Artificial Intelligence Act as appearing to "coerce companies into altering the output of their AI models" (Source: consumerfinancialserviceslawmonitor.com; whitehouse.gov; insideaipolicy.com). The statement extends the Commission's role from consumer-protection enforcement into the federal-state preemption contest documented at AI Federalism.
The statement drew pushback from free-market advocates as of July 6, 2026, with one analyst warning the agency away from a "content policing" role and another cautioning against deception actions built on hard-to-prove consumer expectations about AI "accuracy" (Source: insideaipolicy.com).
The comment period closed on July 31, 2026, having drawn First Amendment and preemption objections. Free Press policy counsel Shilpa Jindia said the Commission is trying to "anoint itself judge and jury over AI content." Ben Winters, director of AI and privacy at the Consumer Federation of America, argued that the FTC Act contains no express preemption provision and that only Congress can override state law. American Association for Justice president John Bey and R Street Institute senior fellows Spence Purnell and Adam Thierer urged withdrawal (Source: broadbandbreakfast.com). Coverage of the comments identifies the authorizing instrument as executive order 14365, signed in December 2025, which directed the Commission to clarify how Section 5 applies to AI models and whether state laws requiring alterations to AI outputs conflict with federal law.
The Consumer Technology Association filed a submission targeting "ambiguity" in the proposed statement, reported August 6, 2026, asking the Commission for a legal safe harbor covering "good-faith" AI design choices and asking Congress to enact a national AI framework that expressly preempts conflicting state laws — an ask that runs past what the Commission itself can grant and into the legislative preemption contest at AI Federalism (Source: insideaipolicy.com). The submission reaches the record through a paywalled lede.
The Center for Democracy and Technology filed an opposing comment, reported August 6, 2026, arguing that the statement would undermine accuracy and security best practices rather than improve them. CDT wrote that the proposal "would not make AI systems more accurate or unbiased" and would instead "empower the Commission to inappropriately pressure companies to shape" model behaviour. The remainder of the filing sits behind the publisher's paywall (Source: insideaipolicy.com). The two August 6 filings are aligned in opposing the statement but differ in what they ask for: CTA seeks a safe harbor and federal preemption, while CDT's objection is that the statement would license the Commission to influence model outputs.
Separately, Sen. Mark Warner (D-VA), the Senate Intelligence Committee's ranking member, unveiled a discussion draft of the AI AGENT Act on June 29, 2026, proposing an FTC-led federal registry of vetted "secure" AI agents, security standards, and fiduciary-style duties for autonomous agents acting for consumers (Source: cyberscoop.com; insideaipolicy.com).
Merger review in AI-adjacent hardware
The Commission cleared IonQ's $1.8 billion acquisition of SkyWater Technology on July 31, 2026 after Chairman Andrew Ferguson and Commissioner Mark Meador could not agree on conditions. Ferguson had proposed requiring IonQ to grant rival quantum-computing firms fair access; Meador said the merger would not lessen competition. The clearance therefore issued from disagreement between the two commissioners rather than from a negotiated remedy (Source: reuters.com).
AI competition work
The FTC maintains an ongoing inquiry into the Microsoft–OpenAI partnership, the most-watched potential AI-antitrust matter (AI Antitrust). As the FTC and DOJ consider reissuing guidance on permissible business collaborations, the American Enterprise Institute filed comments on May 27, 2026 urging the agencies to broaden the antitrust exemption for safety collaborations among frontier developers (Source: insideaipolicy.com).
More than a dozen civil society groups — including the Demand Progress Education Fund, the Consumer Federation of America, and the Institute for Local Self-Reliance — urged the FTC to investigate AI companies for buying, scanning, and destroying books to train models, in a letter that became public on August 21, 2026. The groups ask the agency to determine whether the practice is an unfair method of competition, arguing that a firm doing so is "starving the market" of source materials and that rare books could in some cases be destroyed permanently, with digital firms acquiring the last copies. They do not ask the FTC to restrict model training. The letter itself has not been published (Source: axios.com). See Training Data Walls, AI Antitrust.
Chairman Andrew Ferguson said on September 15, 2026, in what he described as a personal opinion, that "everyone should be deeply suspicious" of AI companies seeking antitrust exemptions while lobbying for regulation, describing the combination as a request for "barriers to entry that will insulate their incumbency from challenge." The remarks responded, without naming it, to Anthropic's call for a narrow antitrust waiver to coordinate a slower pace of development (Source: reuters.com). See AI Antitrust.
Personalized pricing and the sector-specific-enforcement argument (2026)
The FTC opened rulemaking on "personalized pricing" as part of a broader fee-transparency effort initially targeting online food-delivery services. On June 24, 2026 an analyst cited that action as evidence that the proposed federal Sectoral AI Governance Act is unnecessary, arguing that existing sector-specific enforcement by agencies such as the FTC can address AI-enabled harms — including algorithmic pricing — without new cross-cutting AI legislation (Source: insideaipolicy.com). The argument is one position in the recurring debate over whether AI is best governed through existing sectoral regulators applying current authorities or through new AI-specific statutes (see State-Level AI Regulation).
The Commission subsequently prepared a proposed policy statement on personalized pricing under which it intends to enforce its founding statutes and other laws against certain forms of the practice. As of August 20, 2026 the statement had not been issued and was to be released for public feedback. Chairman Ferguson said, "When consumers see a listed price, they expect it to be same price that everyone else sees, not the retailer's estimate of how much they are willing to pay based on their personal data." Only the lede of the report was retrievable (Source: insideaipolicy.com). Earlier and later accounts of the proposal describe different instruments; the discrepancy is set out at Surveillance Pricing. On September 25, 2026 Chairman Ferguson said the agency was preparing to request data from consumer-facing companies for a market study of personalized pricing by merchants themselves; Reuters noted that the earlier study begun under Lina Khan, which she called surveillance pricing, focused on companies offering data and consulting services on pricing rather than on merchants. Ferguson said the FTC has already initiated enforcement actions on personalized-pricing disclosures. The Commission had, on September 24, called for public input on a potential rule to discourage platforms such as Meta and Google from allowing fraudulent or misleading ads (Source: reuters.com).
AI agents and developer liability (2026)
Ferguson said on September 25, 2026 that he would resist treating AI agents as independent actors with "wills and desires of their own," suggesting that developers who instruct agents would be liable for the harm they cause, and that the FTC's authority over companies that fail to disclose data breaches could extend to AI developers (Source: reuters.com). See Andrew Ferguson and AI Liability.
Amazon advertising-auction suit (2026)
On 31 August 2026 the Commission, joined by 22 state attorneys general, sued Amazon in the U.S. District Court for the Western District of Washington over the auctions that set prices for its sponsored-advertising products. The complaint alleges Amazon added an undisclosed surcharge, internally called a "soft reserve price," to a mechanism it described to advertisers as a second-price auction, and estimates the practice likely extracted over $20 billion from 1.2 million advertising customers including more than 500,000 small and medium businesses. Chairman Andrew Ferguson said "Amazon has millions of advertising customers who were misled into paying significantly higher prices" (FTC v. Amazon (advertising auctions)). The theory is one of deception about an automated mechanism rather than about the data feeding a price, which distinguishes it from the personalized-pricing work above.
X Corp consent-order petition (2026)
In May 2026, Elon Musk's X Corp petitioned the FTC to set aside the 2022 Twitter data-privacy consent order "without delay," arguing that Twitter no longer exists as an entity after its mergers into xAI and then SpaceX, and invoking the administration's AI Action Plan. The FTC opened a public comment period on June 3, 2026, running through July 2; early commenters urged denial (Source: arstechnica.com).
Relationships
- related: TAKE IT DOWN Act — FTC-enforced notice-and-takedown regime
- related: AI Companions, Companion Chatbot Harms — Cross-Cutting Analysis, AI and Children — subject matter of the 6(b) inquiry
- related: AI Antitrust — Microsoft–OpenAI inquiry and collaboration-guidance debate
- related: OpenAI, Meta AI, Character.AI, xAI — 6(b) order recipients
- related: Surveillance Pricing — personalized-pricing enforcement and the Section 6(b) market study
- related: FTC Take It Down Act Stakeholder Letter (Ferguson, May 11, 2026) — primary-text compliance guidance