S. 232 — Preventing Algorithmic Collusion Act of 2025

This section establishes definitions for nine terms used in the Act: “antitrust laws” (i.e., as defined in 15 U.S.C. 12(a) and including section 5 of the Federal Trade Commission Act); “commercial terms” (i.e., level of service, availability, output including quantities produced or distributed or amount of service provided, or rebates or discounts); “Commission” (i.e., the Federal Trade Commission); “distribute,” “distribution,” and “distributing” (i.e., selling, licensing, providing access to, or otherwise making available products or services including through subscriptions); “nonpublic competitor data” (i.e., nonpublic data derived from or provided by a competitor in the same or related market, excluding aggregated reports such as industry surveys); “nonpublic data” (i.e., information not widely available to the public, including on price, commercial terms, and related products or services); “person” (i.e., as defined in 15 U.S.C. 12(a)); “price” (i.e., money or other value given in payment for a product or service, including employee or contractor compensation); and “pricing algorithm” (i.e., any computational process, including from machine learning or artificial intelligence, that processes data to recommend or set a price or commercial term in or affecting interstate or foreign commerce).

This section requires any person using or distributing a pricing algorithm to submit to the Attorney General or Federal Trade Commission (FTC), upon written request, a written report on each identified algorithm within 30 days (or later date approved by the requesting authority). The report must cover (1) responsibility for development or distribution; (2) whether the algorithm autonomously sets prices or terms and involves human review; (3) rules or processes for setting or recommending prices or terms; (4) data used, including for training; (5) data sources and collection processes; (6) price discrimination practices and differentiating factors; and (7) changes made prior to certification. A senior executive must certify the report's accuracy under penalty of perjury; all submitted information is confidential as privileged trade secrets exempt from public disclosure under FOIA exemption 4 (5 U.S.C. §552(b)(4)); reports may be shared between the Department of Justice and FTC or with the National Institute of Standards and Technology for technical assistance (with nondisclosure limits); and the provision does not limit other investigative authorities or uses of the information in proceedings.

This section makes unlawful the use or distribution of any pricing algorithm that uses, incorporates, or was trained with nonpublic competitor data, effective 90 days after enactment. The Commission or Attorney General may bring civil actions seeking civil penalties of not less than $10,000 (adjusted for inflation based on the Consumer Price Index) for each day of violation or the sum of the price of each product or service sold using such an algorithm, plus injunctive or other equitable relief.

This section establishes a rebuttable presumption, for purposes of Sherman Act Section 1 (15 U.S.C. 1) and FTC Act Section 5(a) (15 U.S.C. 45(a)), that a defendant entered into an agreement in restraint of trade or engaged in an unfair method of competition through a pricing algorithm if the plaintiff proves that (1) the defendant distributed the algorithm to two or more persons with intent for use (or actual use) to set or recommend prices or terms in the same or related market, or (2) the defendant and at least one other person used it for such purpose in the same or related market. The presumption does not apply if the defendant neither developed nor distributed the algorithm and proves by clear and convincing evidence lack of actual knowledge (or reasonable ability to know) that it used nonpublic competitor data. In applicable civil cases, distributors with knowledge (actual or constructive) that the algorithm used nonpublic competitor data are jointly and severally liable for violations. Subsection (a) takes effect 90 days after enactment.

This section requires any person with $5 million or more in annual revenue that uses a pricing algorithm to recommend or set a price or commercial term to disclose such use—(1) to a customer before purchase of the relevant product or service; and (2) to a current or prospective employee or independent contractor regarding the price or commercial term for services rendered. Disclosures must additionally state, if applicable, whether the algorithm (A) sets or recommends different prices or terms for different customers seeking identical or nearly identical products or services, or for employees or independent contractors providing substantially similar services; or (B) was developed or distributed by a third party, identifying that party. This section deems failure to provide required disclosures an unfair or deceptive act or practice under section 5 of the Federal Trade Commission Act (15 U.S.C. 45), authorizes the Federal Trade Commission to seek civil penalties of not less than $5,000 per day of violation (adjusted for inflation using the Consumer Price Index), and preserves the applicability of antitrust laws.

This section directs the Federal Trade Commission (FTC) to publish, not later than two years after enactment, the results of a study on the use of pricing algorithms. The study must address (1) their prevalence; (2) the frequency of their use to engage in price or wage discrimination; (3) their potential to increase prices, lower wages, reduce output, lower quality, deter innovation, or otherwise harm competition outside price fixing; (4) potential benefits or efficiencies; (5) industries, sectors, or markets warranting additional oversight or regulation to protect competition and consumers; and (6) recommendations for legislation, regulation, or rulemaking on related competition and consumer protection issues.