Existing law governs various business practices in this state, including certain laws relating to the use of technology. Existing law, commonly known as the Cartwright Act, identifies certain acts that are unlawful restraints of trade and unlawful trusts and prescribes provisions for its enforcement through civil actions.

This bill, the Preventing Algorithmic Price Fixing Act, would prohibit a person from selling, licensing, providing, or using a price-setting algorithm, as defined, with the intent or reasonable expectation that it be used by 2 or more competitors, as defined, in the same market if the person knows or should know that the algorithm processes nonpublic data, as defined, to set either: (1) a price or supply level of a good or service or (2) a rent or occupancy level of rental property. The bill would provide a person user of a price-setting algorithm an affirmative defense to liability if the person user exercised reasonable due diligence, as specified. The bill would authorize the Attorney General or a district attorney, city attorney, or county counsel to file a civil action for violations of these provisions, as specified, including for a civil penalty of up to $1,000 per violation, as specified.

Existing law governs various business practices in this state, including certain laws relating to the use of technology.

This bill, the Preventing Algorithmic Price Fixing Act, in addition to any other law, would prohibit a person from selling, licensing, providing, or otherwise providing and a seller, as defined, from using a price-fixing price-setting algorithm, as defined, to with the intent that it be used by 2 or more competitors, as defined, in the same market if the person knows or should know that the algorithm processes nonpublic data, as defined, to set either: (1) a price or supply level of a good or service or (2) a rent or occupancy level of rental property. The bill would provide a seller person an affirmative defense to liability if the seller can show that the seller conducted person exercised reasonable due diligence, as specified. The bill would authorize the Attorney General or a district attorney, city attorney attorney, or county counsel to file a civil action for violations of these provisions, as specified, including for a civil penalty of up to $1,000 per violation, as specified.

Existing law governs various business practices in this state, including certain laws relating to the use of technology.

This bill, the Preventing Algorithmic Price Fixing Act, in addition to any other law, would prohibit a person from selling, licensing, or otherwise providing and a seller, as defined, from using a price-fixing algorithm, as defined, to set either: (1) a price or supply level of a good or service or (2) a rent or occupancy level of rental property. The bill would provide a seller an affirmative defense to liability if the seller can show that the seller conducted due diligence, as specified. The bill would authorize the Attorney General or a city attorney or county counsel to file a civil action for violations of these provisions, as specified, including for a civil penalty of up to $1,000 per violation. violation, as specified.

Existing law governs various business practices in this state, including certain laws relating to the use of technology.

This bill, the Preventing Algorithmic Price Fixing Act, in addition to any other law, would prohibit a business seller, as defined, from using a price-fixing algorithm, as defined, to set either: (1) a price or supply level of a good or service. service or (2) a rent or occupancy level of rental property. The bill would authorize the Attorney General or a city attorney or county counsel to file a civil action for violations of these provisions, as specified, including for a civil penalty of up to $1,000 per violation.

Existing law governs various business practices in this state, including certain laws relating to the use of technology.

This bill, the Preventing Algorithmic Price Fixing Act, would prohibit a business from using a price-fixing algorithm, as defined, to set a price or supply level of a good or service. The bill would authorize the Attorney General or a city attorney or county counsel to file a civil action for violations of these provisions, as specified, including for a civil penalty of up to $1,000 per violation.

Date of Hearing: August 20, 2025

ASSEMBLY COMMITTEE ON APPROPRIATIONS

Buffy Wicks, Chair

SBPCA Bill Id:SB 384

Author:(Wahab) – As Amended Ver:July 17, 2025

Urgency: No State Mandated Local Program: No Reimbursable: No

SUMMARY:

This bill prohibits the sale and use of specified algorithms that process nonpublic input data and are used to set the price or supply level of a good or service, or the rent of occupancy level of rental property.

Specifically, among other provisions, this bill:

1. Prohibits a person from selling, licensing, providing, or using a price-setting algorithm with the intent or reasonable expectation that it be used by two or more competitors in the same market if the person knows or should know that the algorithm processes nonpublic input data to set a price or supply level of a good or service, or a rent or occupancy level of rental property.
2. Provides an affirmative defense to liability for a user of a price-setting algorithm if the user demonstrates, by the preponderance of the evidence, that they exercised reasonable due diligence, as specified.
3. Authorizes enforcement by the Attorney General, a district attorney, a city attorney, or a county counsel, who may file a civil action for actual damages, injunctive relief, restitution, reasonable attorney’s fees and costs, and civil penalties of up to $1,000 per violation.
4. Defines a violation as follows:
5. For a person who sells, licenses, or provides a price-setting algorithm in violation of the bill, each authorized user of, or user under a license for, the algorithm is a separate violation.
6. For a person who uses a price-setting algorithm in violation of the bill, each calendar month of use is a separate violation.

FISCAL EFFECT:

1. Significant costs (Insurance Fund) to the Department of Insurance (CDI). CDI anticipates costs of $196,000 in fiscal year (FY) 2025-26 and $199,000 in FY 2026-27 for additional staffing to manage workload resulting from this bill. CDI explains that insurers are subject to multiple statutory exemptions from antitrust laws. For example, unlike other industries, insurers may act in concert with related insurers on rate setting, participate in certain joint arrangements authorized by law or the insurance commissioner to assure availability of insurance, and share certain data. CDI reports this bill, if applied to insurers, will “upend existing law and industry practice” and “override” these existing antitrust exemptions for the insurance industry. As a result, insurance policies will be voided and insurers will need to file additional forms with CDI, which CDI must process.
2. Costs (Unfair Competition Law Fund) to the Department of Justice (DOJ) to bring enforcement actions as authorized by this bill. Actual costs will depend on the number of enforcement actions pursued by DOJ and the amount of additional work created by each action, but costs may be in the hundreds of thousands of dollars annually. DOJ anticipates costs of $371,000 in fiscal year 2025-26 and $650,000 ongoing annually thereafter for an additional attorney, analyst, and legal secretary in its Consumer Protection Section to handle this workload. DOJ reports it cannot implement the requirements of this bill without an appropriation of additional funds.
3. Cost pressures (Trial Court Trust Fund, General Fund) of an unknown but potentially significant amount to the courts to adjudicate civil actions authorized by this bill. Actual costs will depend on the number of cases filed and the amount of court time needed to resolve each case. It generally costs approximately $1,000 to operate a courtroom for one hour. Although courts are not funded on the basis of workload, increased pressure on the Trial Court Trust Fund may create a demand for increased funding for courts from the General Fund. The fiscal year 2025-26 state budget provides $82 million ongoing General Fund to the Trial Court Trust Fund for court operations.

COMMENTS:

1. Background. Coordinated price-fixing is an anticompetitive business practice prohibited by federal law and the Cartwright Act, California’s primary antitrust statute. The traditional conception of price-fixing involves direct communication by two business competitors who agree to increase their prices in the same market so they each make more money at the expense of consumers. Algorithmic price fixing refers to the use of software to set or recommend prices in ways that result in coordinated outcomes between competitors without any formal agreement. In the contemporary economy, many businesses use algorithms that process huge quantities of data and make recommendations about the prices a business should set. Depending on what data is processed by an algorithm, how many businesses are using the same algorithm in the same market, and other factors, these algorithmic tools can create the same result as coordinated price-fixing: businesses setting similarly high prices to the detriment of consumers. However, the business community reports that use of price-setting algorithms is ubiquitous and not inherently collusive.
2. Related Legislation. This bill is one of several this session that tries to define when use and distribution of price-setting algorithms becomes an anticompetitive practice tantamount to price-fixing. SB 295 (Hurtado) is similar in scope and approach to this bill: both bills prohibit the distribution and use of algorithms that process nonpublic data and are used by two or more competitors in the same market to set or recommend prices. Both bills cover roughly the same scope of goods or products, services, and rental property. Both bills include an affirmative defense to liability for an algorithm user who demonstrates they exercised due diligence before using the algorithm, including by obtaining assurances from the distributor that the algorithm does not process the kind of data prohibited by each bill. Both bills authorize enforcement by public prosecutors, void contract terms that conflict with the bill’s provisions, and state they do not limit the applicability of other antitrust laws.

However, there are some differences between the bills. For example, the bills authorize different civil penalties: the maximum penalty under SB 295 is $25,000 per violation, while the maximum penalty under SB 384 is $1,000 per violation. In addition to violations for use and distribution of specified algorithms, SB 295 makes it unlawful for a person to make a recommendation based on the use of a pricing algorithm that violates the bill’s terms. SB 384 does not contain a parallel violation. The bills also use different definitions for the algorithms governed by each bill, and slightly different definitions of the type of data that renders use of an algorithm unlawful.

Other bills in this area include SB 52 (Perez), which prohibits specified uses of rental pricing algorithms that process nonpublic competitor data. SB 52 is pending in this committee. AB 325 (Aguiar-Curry) specifies that using or distributing a common pricing algorithm to further a price-fixing conspiracy is a violation of the Cartwright Act. AB 325 is pending on the Senate Appropriations Committee’s suspense file.

Analysis Prepared by: Annika Carlson / APPR. / (916) 319-2081

Date of Hearing: July 16, 2025

Fiscal:

ASSEMBLY COMMITTEE ON PRIVACY AND CONSUMER PROTECTION

Rebecca Bauer-Kahan, Chair

SBPCA Bill Id:SB 384

Author:(Wahab) – As Amended Ver:June 26, 2025

SENATE VOTE: 28-10

PROPOSED AMENDMENTS

SUBJECT: Preventing Algorithmic Price Fixing Act: prohibition on certain price-setting algorithm uses

SYNOPSIS

Antitrust laws have been on the books for over a century and have evolved along with the means of reaching unlawful price-fixing agreements. As smoke-filled backrooms and handshakes gave way to the phone, fax, pager, and email, the law has adapted to address new challenges. The new frontier is algorithmic price-fixing, which can attain uncompetitive ends with greater efficiency and scale – and yet is even more difficult to detect, let alone prove in court.

While algorithmic collusion appears to be on the rise in certain sectors of the economy, competition in the legislative marketplace of ideas is thriving. Before the Committee today are three overlapping bills – this bill, SB 52 (Pérez), and SB 295 (Hurtado) – that, in similar ways, seek to prohibit the distribution and use of algorithmic price-fixing tools that process nonpublic competitor data. Whereas SB 52 focuses specifically on housing, the other two bills apply more broadly: SB 295 applies to products, services, and rental property, and this bill applies to goods, services, and rental housing

Specifically, this bill prohibits a person from selling, licensing, providing, or using a price-setting algorithm – which by definition processes nonpublic competitor data – with the intent that it be used by two or more competitors in the same market if the person knows or should know that the algorithm processes nonpublic input data to set either 1) a price or supply level of a good or service, or 2) a rent or occupancy level of rental property. Public prosecutors may bring civil actions against violators.

This measure is author-sponsored and supported by a broad coalition, including public interest and housing advocates. It is opposed by an industry coalition led by the Chamber of Commerce, among others.

Committee amendments described in Comment # 8 include “reasonable expectation” in addition to intent in the provision described above. The amendments also lower the threshold for an affirmative defense and provide that the bill does not apply to the extent that nonpublic competitor data used by the price-setting algorithm is more than one year old, overrides conflicting contractual provisions, and does not limit the applicability of antitrust laws.

The bill passed the Judiciary Committee on a 9-2 vote.

THIS BILL:

1. Prohibits a person from selling, licensing, providing, or using a price-setting algorithm with the intent that it be used by two or more competitors in the same market if the person knows or should know that the algorithm processes nonpublic input data to set either of the following:
   1. A price or supply level of a good or service.
   2. A rent or occupancy level of rental property.
2. Provides that it is an affirmative defense to liability if the person demonstrates by clear and convincing evidence that they exercised due diligence, as specified.
3. Provides that, with respect to distribution in violation of the bill, each authorized user constitutes a separate violation. With respect to use in violation of the bill, provides that each calendar month of use constitutes a separate violation.
4. Provides that public prosecutors may file a civil action for injunctive relief, restitution, civil penalties of up to $1,000, and reasonable attorney’s fees and costs.
5. Defines, among other terms:
   1. “Price-setting algorithm” as a software, computer system, computer process, algorithmic program, or artificial intelligence that processes nonpublic input data for the purpose of producing a pricing or rental strategy. Excludes from this definition “multiple listing services,” as defined in existing law.
   2. “Nonpublic input data” as data that is confidential, nonpublic, and sensitive information of competitors.
   3. “Competitors” as two or more persons or business entities, including landlords, that offer similar or substitutable goods, services, or real property for lease in the same relevant market to the same or overlapping customer base.

EXISTING LAW:

1. Under the federal Sherman Antitrust Act of 1890, prohibits any contract, combination in the form of trust or otherwise, or conspiracy, that unreasonably restrains trade. (15 U.S.C. § 1.) Prohibits monopolizing or attempting to monopolize, or conspiring to monopolize, trade or commerce. (15 U.S.C. § 2.)
2. Under California’s Cartwright Act, makes every “trust” unlawful, against public policy, and void. (Bus. & Prof. Code § 16726.) Defines a “trust” as a combination of capital, skill or acts by two or more persons in order to do any of the following:
   1. Create or carry out restrictions in trade or commerce.
   2. Limit or reduce the production, or increase the price of merchandise or of any commodity.
   3. Prevent competition in manufacturing, making, transportation, sale or purchase of merchandise, produce or any commodity.
   4. Fix at any standard or figure, whereby its price to the public or consumer is in any manner controlled or established, any article or commodity of merchandise, produce or commerce intended for sale, barter, use or consumption in California.
   5. Make or enter into or execute or carry out any contracts, obligations or agreements of any kind or description, by which they do all or any or any combination of any of the following:
      1. Bind themselves not to sell, dispose of or transport any article or any commodity or any article of trade, use, merchandise, commerce or consumption below a common standard figure, or fixed value.
      2. Agree in any manner to keep the price of such article, commodity or transportation at a fixed or graduated figure.
      3. Establish or settle the price of any article, commodity or transportation between them or themselves and others, so as directly or indirectly to preclude a free and unrestricted competition among themselves, or any purchasers or consumers in the sale or transportation of any such article or commodity.
      4. Agree to pool, combine or directly or indirectly unite any interests that they may have connected with the sale or transportation of any such article or commodity, that its price might in any manner be affected. (Bus. & Prof. Code § 16720.)
3. Provides that the Attorney General and county district attorneys may bring criminal or civil actions to enforce the Cartwright Act. Corporations are subject to fines of up to $1 million; individuals are subject to fines of up to $250,000 and imprisonment for three years. (Bus. & Prof. Code § 16755.)
4. Grants a private right of action in which plaintiffs may recover treble damages, injunctive relief, costs, attorney’s fees, and interest on actual damages. (Bus. & Prof. Code §§ 16750, 16761.)
5. Defines a “multiple listing service” (MLS) as “a facility of cooperation of agents and appraisers, operating through an intermediary that does not itself act as an agent or appraiser, through which agents establish express or implied contracts for compensation between agents that are MLS participants in accordance with its MLS rules with respect to listed properties in a listing agreement, or that may be used by agents and appraisers, pursuant to the rules of the service, to prepare market evaluations and appraisals of real property.” (Civ. Code § 1087.)

COMMENTS:

1) Author’s statement. According to the author:

Rapid technological advances have resulted in corporations using tools and resources that result in digital collusion. Corporations rely on the existence of these tools as permission to use them without considering the legal implications. SB 384 places responsibility on these corporations to evaluate the tools they use to process data associated with their own pricing structures and stock by prohibiting the use of tools that collect private data from two or more corporations for the purposes of analyzing and processing the data to create pricing models. This prohibition will force corporations to be more discerning regarding the digital tools they use, and reduce the likelihood of corporations accidentally engaging in digital handshakes that result in price-fixing and market manipulation.

2) Antitrust laws. Two closely related antitrust laws – the federal Sherman Act and the state’s Cartwright Act – are implicated in any form of price-fixing.

Sherman Act. Section 1 of the federal Sherman Act prohibits concerted action that restrains trade, while Section 2 covers concerted action and independent action, but “only when it threatens actual monopolization,” a higher bar than restraint of trade.^[1] According to the United States Supreme Court:

The reason Congress treated concerted behavior more strictly than unilateral behavior is readily appreciated. Concerted activity inherently is fraught with anticompetitive risk. It deprives the marketplace of the independent centers of decisionmaking that competition assumes and demands. In any conspiracy, two or more entities that previously pursued their own interests separately are combining to act as one for their common benefit. This not only reduces the diverse directions in which economic power is aimed but suddenly increases the economic power moving in one particular direction. Of course, such mergings of resources may well lead to efficiencies that benefit consumers, but their anticompetitive potential is sufficient to warrant scrutiny even in the absence of incipient monopoly.^[2]

“The relevant inquiry” under section 1 “is whether there is a ‘contract, combination . . . , or conspiracy’ amongst ‘separate economic actors pursuing separate economic interests, such that the agreement ‘deprives the marketplace of independent centers of decisionmaking,’ and therefore of ‘diversity of entrepreneurial interests,’ and thus of actual or potential competition.”^[3] In other words, “The ‘crucial question’ prompting Section 1 liability is ‘whether the challenged anticompetitive conduct ‘stems from [lawful] independent decision or from an agreement, tacit or express.’”^[4]

Cartwright Act. “[B]roader in range and deeper in reach”^[5] than its federal counterpart, California’s Cartwright Act (Act) “‘generally outlaws any combinations or agreements which restrain trade or competition or which fix or control prices.’”^[6] The Act is “premised on the notion that competition yields efficient resource allocation, lower prices, higher quality, and greater social welfare.”^[7] “‘At its heart is a prohibition against agreements that prevent the growth of healthy, competitive markets for goods and services and the establishment of prices through market forces.’”^[8] “The [A]ct’s principal goal is the preservation of consumer welfare.”^[9]

Under the Act, a violation requires “a combination of capital, skill or acts by two or more persons” that seeks to achieve an anticompetitive end.^[10] A complaint pursuant to the Act must allege: “(1) the formation and operation of the conspiracy, (2) the wrongful act or acts done pursuant thereto, and (3) the damage resulting from such act or acts.”^[11]

Concerted action. “Two forms of conspiracy may be used to establish a violation of antitrust laws: a horizontal restraint, consisting of a collaboration among competitors; or a vertical restraint, based upon an agreement between business entities occupying different levels of the marketing chain.”^[12] A hybrid of horizontal and vertical agreements is sometimes referred to as a “hub-and-spoke” conspiracy, in which a central actor, or hub, enters into vertical agreements with spokes, such as competing manufacturers or distributors. If the spokes have horizontal agreements with each other, the conspiracy is “rimmed”, whereas if they do not, it is a “rimless” hub-and-spoke.

Certain types of agreements that restrain trade are illegal per se because they almost always undermine competition, while others are subject to a “rule of reason” review, which requires the plaintiff to show that the agreement harms competition more than it helps.^[13] Most horizontal agreements are per se violations,^[14] whereas vertical agreements are usually analyzed under the rule of reason.^[15] Price fixing, however, is per se illegal regardless of whether it occurs between competitors or businesses at different economic levels.^[16]

On the other hand, merely exchanging information, including about prices, is not itself illegal unless it is part of an express or tacit agreement to fix prices.^[17] Agreements “may be inferred on the basis of conscious parallelism, when such interdependent conduct is accompanied by circumstantial evidence and plus factors.”^[18] Plus factors can include “a common motive to conspire, evidence that shows that the parallel acts were against the apparent individual economic self-interest of the alleged conspirators, and evidence of a high level of interfirm communications.”^[19]

3) Price-fixing algorithms. Pricing algorithms are commonly used to help set prices that are responsive to market conditions, which may increase market efficiency in competitive markets. However, they can also facilitate price-fixing, thereby decreasing market efficiency and hampering competition.

Algorithmic collusion is not new. In 1992, the United States Department of Justice (DOJ) filed suit against eight of the nation’s largest airlines in connection with an algorithmic pricing system, known as the Airline Tariff Publishing Company (ATP), which was used to increase the cost of airplane tickets by potentially upwards of a billion dollars during a four-year period. “By supplying or withdrawing changes in fares, the airlines told each other what fares they wanted to charge in which markets, what competitors’ fares were acceptable to them, and what deals they were willing to make.” ^[20] The attorney in charge of DOJ’s Antitrust Division stated, ‘The airlines used the ATP fare dissemination system to carry on conversations just as direct and detailed as those traditionally conducted, by conspirators over the telephone or in hotel rooms. Although their method was novel, their conduct amounted to price fixing, plain and simple.’ Two of the airlines entered a consent decree and the other six entered into a settlement with the DOJ.^[21]

The rise of machine-learning pricing algorithms has intensified concerns about anti-competitive behavior, particularly tacit collusion.^[22] Unlike older rule-based systems, modern algorithms can rapidly assimilate market data, predict demand fluctuations, and adjust prices based on competitor behavior, often reinforcing strategies that maximize profits in an anticompetitive fashion.^[23] In particular, models that use reinforcement learning – a training process that uses rewards and punishments to orient a model’s behavior towards attaining a specific goal^[24] – and real-time data feedback loops can adapt to function in a manner that sustains high prices, effectively facilitating tacit collusion without explicit human agreement.^[25] Moreover, algorithms’ faster response times and improved demand predictions may help firms sustain collusive pricing structures by swiftly detecting and punishing deviations, leading to supra-competitive prices.^[26]

With respect to the impact of pricing algorithms in the housing context, in 2024 the White House issued a report concluding as follows: “We find that anticompetitive pricing costs renters in algorithm-utilizing buildings an average of $70 a month. In total, we estimate the costs to renters in 2023 was $3.8 billion. This estimate is likely a lower bound on the true costs.”^[27]

4) AI collusion cases. A number of pending federal cases allege that the use of a common pricing algorithms violates the Sherman Act. Some key examples follow.

RealPage. In October of 2022, ProPublica published an investigation of RealPage’s rental housing pricing algorithm. This popular software, used by many of the largest property managers who control thousands of apartments in metropolitan areas throughout the country, collects information from the property managers, including private lease transactions and occupancy data, that is then fed into a common algorithm that recommends optimal rental rates.^[28] This led to numerous class-action lawsuits against RealPage, as well as a lawsuit by Attorney General Rob Bonta, along with the DOJ and eight other attorneys general. The litigation is ongoing.^[29]

In a filing with the court, the DOJ set forth its view that “[a]s with other actions taken in concert, competitors’ joint use of common algorithms can remove independent decision making. . . . Put another way, whether firms effectuate a price-fixing scheme through a software algorithm or through human-to-human interaction should be of no legal significance. Automating an anticompetitive scheme does not make it less anticompetitive.” The DOJ continued:

The question in this case is whether the defendants have violated Section 1 of the

Sherman Act by allegedly knowingly combining their sensitive, nonpublic pricing and supply information in an algorithm that they rely upon in making pricing decisions, with the knowledge and expectation that other competitors will do the same. Although not every use of an algorithm to set price qualifies as a per se violation of Section 1, taking the allegations set forth in the complaints as true, the alleged scheme meets the legal criteria for per se unlawful price fixing.^[30]

RENTmaximizer. Another pending case similarly involves allegations that competing landlords violated section 1 of the Sherman Act by, among other things, unlawfully agreeing to use a centralized pricing algorithm to artificially inflate multifamily rental prices.^[31] The DOJ argued that “competitors’ jointly delegating key aspects of their decisionmaking to a common algorithm” amounts to per se concerted action “because doing so ‘joins together separate decisionmakers’ and thus ‘deprives the marketplace of independent centers of decisionmaking.’”^[32] Furthermore, “[w]here, as here, plaintiffs’ allegations involve a conspiracy to centralize pricing decisions in a third-party algorithm, it is irrelevant to the scheme whether landlords share confidential information among themselves or with only the pricing agent; the alleged scheme is designed to obviate the need for competitors to share information directly with each other.”^[33]

Cendyn. A federal district court recently dismissed a class action lawsuit alleging that Las Vegas hotel operators engaged in illegal price-fixing by using Cendyn’s revenue management software.^[34] The court highlighted that the pricing recommendations were not based on nonpublic, competitively sensitive information; rather, it was public information available from online listings and travel agencies. The court also concluded that the plaintiffs failed to allege that hotel operators “agreed to be bound by [Cendyn’s] pricing recommendations, much less that they all agreed to charge the same prices.”^[35] The decision is on appeal to the Ninth Circuit. The DOJ has argued that “an invitation for collective action followed by conduct showing acceptance” such as “joint use” of the pricing algorithm amounts to concerted action, which may be a per se violation if the algorithm sets a “default or starting point price,” if the hotel ultimately charges a different price.^[36] DOJ also argued that the lower court erred by failing to recognize allegations supporting the conclusion that “Cendyn combines non-public, competitively sensitive information to generate prices.”^[37]

A similar case involving use of Cendyn’s pricing software by Atlantic City hotels was also dismissed. Although the DOJ and the Federal Trade Commission argued usage of pricing algorithms is unlawful even when co-conspirators retain pricing discretion and do not communicate directly with each other, the court found that the alleged co-conspirators used the pricing algorithm at different points in time, no confidential or otherwise nonpublic information was exchanged, and the alleged co-conspirators were not bound to accept the algorithm’s pricing recommendations.^[38]

Potato cartel. From July 2022 to 2024, potato prices spiked nearly 50%.^[39] A pending 2024 class action alleges the four largest potato merchants – which together control 95% of the potato market – used potato price data aggregation services, collective action through trade associations, and direct communication to exchange pricing data and other sensitive information, leading potato prices to spike by nearly 50% between July 2022 and 2024.^[40] Among other things, the complaint alleges, “Defendants and their co-conspirators entered into continuing agreement to regularly exchange detailed, timely, competitively sensitive information which is non-public and is about their operations. This agreement is an unreasonable restraint of trade under the Sherman Act.”^[41]

Multiplan. A recent lawsuit alleged that Multiplan – a healthcare technology company – colluded with insurers, including Aetna, Cigna, UnitedHealth, and Blue Cross Blue Shield Association, to suppress reimbursement rates using its pricing algorithm. In ruling on the defendants’ motion to dismiss, the court found that the plaintiffs failed to allege sufficient facts showing Multiplan’s algorithm compiles competitively sensitive data from insurers; however, the plaintiffs had sufficiently alleged that Multiplan itself “played an active role as a go-between” for third-party payors.^[42]

In sum, these cases and the arguments advanced by DOJ appear to generally emphasize the nonpublic, competitively sensitive nature of pricing information shared via an algorithmic intermediary, as well as the extent to which the competitors have discretion to depart from the recommended prices generated by the algorithm.

5) Law Revision Commission working group report. AR 95 (Cunningham, 2021) called upon the California Law Revision Commission to study whether the Cartwright Act requires updating. To assist in its study, the Commission formed working groups of experts, one of which issued a report on “Competition and Artificial Intelligence.” Regarding algorithmic collusion, the working group concludes:

. . . The Cartwright Act generally prohibits any combinations or agreements which unreasonably restrain trade or fix or control prices. As currently interpreted by the courts, the Cartwright Act requires a “combination” or “concerted action” between 2 or more independent economic entities. Given the increasing use of software programs containing or relying on pricing algorithms, the Legislature might consider declaring that the “concerted action” requirement of the Cartwright Act encompasses multiple competitors that knowingly use the same or similar revenue management software programs containing or relying on pricing algorithms that utilize nonpublic competitor information to train or inform any price recommendations.

Consistent with the position of the DOJ . . ., the Legislature might also clarify that direct communications are not required to show proof of a “combination” or “concerted action” among competitors, as the Cartwright Act covers tacit as well as express agreements. This is in accord with the position of the DOJ . . . that Section 1 of the Sherman Act prohibits “tacit agreements”—that is where one co-conspirator invites participation in an illegal price-fixing scheme and other co-conspirators act in accordance with the scheme, showing acceptance through a course of conduct.

Further, the Legislature might make clear that the Act prohibits competitors from “delegating key aspects of pricing decision making to a common entity, even if the competitors never communicate with each other directly.” Further, to refute the argument that there can be no actionable claim of price fixing because the algorithm’s recommendations are not binding, the Legislature could declare that, under the Cartwright Act, “an agreement among competitors to fix the starting point of pricing is per se unlawful, no matter what prices the competitors ultimately charge.” ^[43]

6) What this bill would do. This bill prohibits a person from selling, licensing, providing, or using a price-setting algorithm – which by definition processes nonpublic competitor data – with the intent that it be used by two or more competitors in the same market if the person knows or should know that the algorithm processes nonpublic input data to set either 1) a price or supply level of a good or service, or 2) a rent or occupancy level of rental property. Public prosecutors may bring civil actions against violators. A defendant may assert, as an affirmative defense to liability, that they exercised due diligence, which must be demonstrated by clear and convincing evidence.

Proponents argue that traditional antitrust law’s focus on the presence of some form of agreement fails to clearly address algorithmically-mediated price coordination that occurs without a collusive meeting of the minds. This bill is necessary, they assert, to combat this growing phenomenon, and thereby help reduce soaring prices.

In describing how the bill would apply, the Assembly Judiciary Committee analysis of the bill set forth the following hypothetical scenarios:

Hypothetical #1 – Lawful under SB 384. A permissible example under the bill would be a travel app such as Hopper or Kayak that aggregates publicly available airfare data from airlines to help consumers identify the best time to purchase tickets. These apps do not rely on nonpublic or competitively sensitive information, nor do they communicate price recommendations back to the airlines. Their use is unilateral and consumer-facing, promoting price transparency rather than coordination. Therefore, such conduct would not be covered by SB 384.

Hypothetical #2 – Unlawful under SB 384. Suppose a dynamic pricing vendor licenses an AI-powered tool to three large grocery chains in the same regional market. The algorithm collects nonpublic pricing data, discounting practices, and volume information from each chain’s internal systems, and uses that data to recommend “market-optimized” pricing strategies across all three. If the vendor knows or reasonably expects that the algorithm will be used to influence pricing decisions across these competitors, and if the grocery chains know or should know the algorithm is trained on confidential data and used across the market, then both the vendor and users could be liable under SB 384.^[44]

7) Related legislation. Several legislative proposals that seek to address algorithmic collusion have recently been proposed, including:

• At the federal level, Senator Amy Klobuchar recently reintroduced her “Preventing Algorithmic Collusion Act” (S. 232) which makes it presumptively unlawful for a person to use or distribute a pricing algorithm that uses, incorporates, or was trained with nonpublic competitor data.
• Colorado Governor Jared Polis recently vetoed a bill that would have prohibited the sale, distribution, or use of algorithmic devices using public and nonpublic data to coordinate the setting of price or occupancy levels between two or more landlords, a violation of which would be deemed a violation of the Colorado Antitrust Act of 2023. In his veto message, Governor Polis cited concerns around inadvertently banning tools that may assist in efficiently managing residential real estate to ensure people can access housing. He stated he would open to supporting a bill that “makes a distinction between collusive and non-collusive uses of nonpublic competitor data.”
• AB 325 (Aguiar-Curry, 2025) provides that it is unlawful to use or distribute a common pricing algorithm as part of a contract, combination in the form of a trust, or conspiracy to restrain trade or commerce in violation of the Cartwright Act. The bill also provides that it is unlawful for a person to use or distribute such algorithms if the person coerces another to set or adopt a recommended price or commercial term for the same or similar products or services. The bill additionally changes the pleading standard in Cartwright Act cases. The bill is pending in the Senate Appropriations Committee.
• SB 52 (Pérez) is similar to this bill, but it applies to rental housing only.
• SB 295 (Wahab, 2025) is similar to this bill, but it applies to products rather than goods, and the civil penalty is up to $25,000 per violation.

8) Amendments. The author has agreed to amend the bill as follows:

22949.85. (a) A person shall not sell, license, provide, or use a price-setting algorithm with the intent or reasonable expectation that it be used by two or more competitors in the same market if the person knows or should know that the algorithm processes nonpublic input data to set either of the following:

(1)  A price or supply level of a good or service.

(2) A rent or occupancy level of rental property.

(b) It shall be an affirmative defense to liability for a person user of a price-setting algorithm who demonstrates by clear and convincing evidence a preponderance of the evidence that they exercised reasonable due diligence, including obtaining written assurances from the person selling, licensing, or providing the algorithm that the algorithm does not process nonpublic input data.

(c) (1) For a person who sells, licenses, or provides a price-setting algorithm in violation of subdivision (a), each authorized user of, or user under a license for, the price-setting algorithm constitutes a separate violation.

(2) For the purposes of a person who uses a price-setting algorithm in violation of subdivision (a), each calendar month of use constitutes a separate violation.

(d) The Attorney General or a district attorney, in the name of the people of the State of California, or a city attorney or county counsel, in the name of the city or county, may file a civil action for a violation of this section to recover actual damages or for injunctive relief, restitution, or civil penalties of up to one thousand dollars ($1,000) per violation, or any combination of those remedies. The court shall award reasonable attorney’s fees and costs to the Attorney General, district attorney, city attorney, or county counsel, as applicable, if they are the prevailing party in the action.

(e) For the purposes of this section:

(1) “Antitrust laws” has the same meaning as defined in the Clayton Act (15 U.S.C. Sec. 12), and includes Section 45 of Title 15 of the United States Code, including provisions commonly known as the Cartwright Act (Chapter 2 (commencing with Section 16700)).

(2) “Artificial intelligence” means an engineered or machine-based system that varies in its level of autonomy and that can, for explicit or implicit objectives, infer from the input it receives how to generate outputs that can influence a physical or virtual environment.

(23) “Nonpublic input data” means data that is confidential, nonpublic, and sensitive information of competitors, but does not include data was collected more than one year before the use or distribution of the price-setting algorithm.

(34) (A) “Price-setting algorithm” means a software, computer system, computer process, algorithmic program, or artificial intelligence that processes nonpublic input data for the purpose of producing a pricing or rental strategy.

(B) “Price-setting algorithm” does not include a multiple listing service, as that term is defined in Section 1087 of the Civil Code.

(45) “Competitors” means two or more persons or business entities, including landlords, that offer similar or substitutable goods, services, or real property for lease in the same relevant market to the same or overlapping customer base.

(f) A contract that conflicts with this section is to that extent void.

(g) This section does not limit the applicability of antitrust laws.

ARGUMENTS IN SUPPORT: Housing California writes:

Artificial intelligence (AI) companies are using algorithms to set rent prices based on rental data from thousands of landlords and other sources. These AIdriven rent-setting algorithms turn competitors into collaborators, facilitating an unlawful information-sharing operation that manipulates the housing market for an anti-competitive gain. This is tech-powered exploitation is worsening the already dire affordability crisis. President Biden’s Council of Economic Advisors (CEA) has found that, on average, this practice costs renters at least $3.8 billion in buildings that use these algorithms to determine rents. RealPage, the largest company providing these pricing algorithms, controls 80% of the commercial revenue management market. It is estimated that their algorithms allow landlords to increase rents by 3-7% above what is considered competitive.

Oakland Privacy writes:

Senate Bill 384 would eliminate such use of algorithms to price-fix in the above-described manner and prohibit their use in California, although potentially such algorithmic programs could be used or trained using public data to develop pricing strategies, and that use would remain untouched by this bill. We support this general concept as the adaption of classic antitrust law to the same old price-fixing behavior in new guises and formats due to technological innovation and the increasing capacity of artificial intelligence programming. Consumers are entitled to the benefits of marketplace competition between various sellers, which is the primary lever for downward pressure on prices in pure market-based capitalism. When sellers collude secretively to keep prices artificially high, then consumers lose their strongest weapon which is the ability to walk to another seller should one particular seller raise their price too high.

ARGUMENTS IN OPPOSITION: A coalition of industry opponents, led by Chamber of Commerce, argues:

Like several other pricing algorithm bills moving through the Legislature, SB 384 appears rooted in an assumption that pricing algorithms are inherently problematic or unlawful, as opposed to attempting to identify and halt demonstrably anti-competitive behaviors or price-fixing practices. Pricing algorithms are, in fact, extremely common tools that enable businesses to reduce costs, improve efficiency by avoiding manual pricing, dynamically adjust prices in response to market conditions and changes in supply and demand (including to decrease prices), and ultimately reduce costs to consumers as well – and they can do so without involving any anticompetitive conduct.

In contrast, price collusion (or price fixing) is problematic and is clearly illegal under current federal and state laws. Indeed, existing antitrust laws prohibit competitors from colluding on pricing any manner, whether through using a pricing algorithm or otherwise. In other words, whether a price-fixing conspiracy is hatched by salespeople conspiring or computers running algorithms, collusion is collusion and is already covered effectively by existing law. To be clear, however, the use of a pricing algorithm does not inherently constitute price fixing.

Retailers use pricing algorithms to ensure they are offering the most competitive prices to consumers. Realtors use them to help clients set home prices. Banks use them to set terms (e.g. rates and fees) for services. Hospitality, airlines, transportation network companies, utilities, ticket venues, and many others use them for dynamic pricing. The list goes on.

Thus, legislation such as SB 384 risks merely removing a valuable tool for setting dynamic pricing and impose significant costs on all businesses who use price algorithms – especially smaller businesses – thereby reducing competition, rather than promoting it. In the end, this hurts not only businesses, taking them back to pre-technological times, but it also hurts consumers, effectively doing away with price-comparison shopping and competitive/dynamic pricing by businesses. (Emphasis in original.)

REGISTERED SUPPORT / OPPOSITION:

Support

Aids Healthcare Foundation
California Housing Partnership
California Rural Legal Assistance Foundation, INC.
California School Employees Association
Housing California
Oakland Privacy
Open Markets Institute

Oppose

American Property Casualty Insurance Association
Association of National Advertisers
Calbroadband
California Apartment Association
California Chamber of Commerce
California Credit Union League
California Fuels and Convenience Alliance
California Hospital Association
California Hotel & Lodging Association
California Restaurant Association
California Retailers Association
California Travel Association
Civil Justice Association of California (CJAC)
Insights Association
National Association of Mutual Insurance Companies
Personal Insurance Federation of California
Software Information Industry Association
Technet

Oppose Unless Amended

Insurance Services Office, INC.
Realpage, INC.

Analysis Prepared by: Josh Tosney / P. & C.P. / (916) 319-2200

1. Copperweld Corp. v. Independence Tube Corp. (1984) 467 U.S. 752, 767. ↑

2. Id. at pp. 768-769. ↑

3. Am. Needle, Inc. v. NFL (2010) 560 U.S. 183, 195, citations omitted. ↑

4. In re Dynamic Random Access Memory (DRAM) Indirect Purchaser Antitrust Litig. (9th Cir. 2022) 28 F.4th 42, 46, quoting Bell Atl. Corp. v. Twombly (2007) 550 U.S. 544, 553. ↑

5. In re Cipro Cases I & II (2015) 61 Cal.4th 116, 160-161. However, the Cartwright Act does not prohibit unilateral conduct. ↑

6. Pacific Gas & Electric Co. v. County of Stanislaus (1997) 16 Cal.4th 1143, 1147. ↑

7. Ahn v. Stewart Title Guaranty Co. (2023) 93 Cal.App.5th 168, 179. ↑

8. Ibid. ↑

9. In re Cipro Cases I & II, supra, 61 Cal. 4th at p. 136. ↑

10. Bus. & Prof. Code § 16720. ↑

11. Smith v. State Farm Mut. Auto. Ins. Co. (2001) 93 Cal. App. 4th 700, 722; Asahi Kasei Pharma Corp. v. CoTherix, Inc. (2012) 204 Cal.App.4th 1, 8; Olean Wholesale Grocery Coop., Inc. v. Bumble Bee Foods LLC (9th Cir. 2022) 31 F.4th 651, 665, fn. 8. ↑

12. G.H.I.I. v. MTS, Inc. (1983) 147 Cal.App.3d 256, 267, emphasis in original. ↑

13. People v. Bldg. Maint. Contractors’ Ass’n (1957) 41 Cal. 2d 719, 727. ↑

14. See Knevelbaard Dairies v. Kraft Foods, Inc. (9th Cir. 1979) 232 F.3d 979, 986. ↑

15. In re Musical Instruments & Equip. Antitrust Litig. (9th Cir. 2015) 798 F.3d 1186 n.3; United States v. Joyce (9th Cir. 2018) 895 F.3d 673, 677. ↑

16. Mailand v. Burckle (1978) 20 Cal.3d 367, 377. ↑

17. Aguilar v. Atl. Richfield Co. (2001) 25 Cal. 4th 826, 862-863 (“‘Competition does not become less free merely because the conduct of commercial operations becomes more intelligent through the free distribution of knowledge of all the essential factors entering into the commercial transaction’”). ↑

18. In re Automobile Antitrust Cases I & II (2015) 1 Cal.App.5th 127, 169, citations and nested quotation marks omitted. ↑

19. Ibid. ↑

20. “Justice Department Settles Airlines Price Fixing Suit, May Save Consumers Hundreds of Millions of Dollars” (1994), https://www.justice.gov/archive/atr/public/press_releases/1994/211786.htm. ↑

21. Ibid. ↑

22. Clark et al, “Pricing Algorithms as Third-Party Facillitators of Collusion” American Bar Association (Dec. 2024), p. 3, https://www.americanbar.org/content/dam/aba/publications/antitrust/source/2024/december/pricing-algorithms-third-party-facilitators-collusion.pdf. ↑

23. Ibid. ↑

24. Mummert et al., “What is reinforcement learning?” IBM Developer (September 15, 2022), developer.ibm.com/learningpaths/get-started-automated-ai-for-decision-making-api/what-is-automated-ai-for-decision-making. ↑

25. “Pricing Algorithms as Third-Party Facillitators of Collusion,” supra, pp. 3-5. ↑

26. Ibid. ↑

27. White House, The Cost of Anticompetitive Pricing Algorithms in Rental Housing (Dec. 17, 2024), https://bidenwhitehouse.archives.gov/cea/written-materials/2024/12/17/the-cost-of-anticompetitive-pricing-algorithms-in-rental-housing/. ↑

28. Heath Vogell, “Rent Going Up? One Company’s Algorithm Could Be Why,” ProPublica (Oct. 15, 2022), https://www.propublica.org/article/yieldstar-rent-increase-realpage-rent. ↑

29. See In re RealPage, Inc., Rental Software Antitrust Litig. (No. II) (M.D.Tenn. 2023) 709 F. Supp. 3d 478, 492. ↑

30. Memorandum of Law in Support of Statement of Interest of the United States, In re RealPage, Case No. 3:23-MD-3071 (M.D. Tenn Nov. 15, 2023), https://www.justice.gov/d9/2023-11/418053.pdf. (Emphasis added.) ↑

31. See Duffy v. Yardi Sys., Inc. (W.D.Wash. Dec. 5, 2024, No. 2:23-cv-01391-RSL) 2024 U.S.Dist.LEXIS 220641. ↑

32. Statement of Interest (March 1, 2024), in Yardi, supra, pp. 2-3, https://www.justice.gov/d9/2024-03/420301.pdf. ↑

33. Id. at p. 7, fn. 4. (Emphasis in original.) ↑

34. Gibson v. Cendyn Grp., LLC (D.Nev. May 8, 2024, No. 2:23-cv-00140-MMD-DJA) 2024 U.S.Dist.LEXIS 83547. ↑

35. Id. at 6. ↑

36. Brief for the DOJ as Amicus Curiae, Gibson v. Cendyn Group LLC, No. 24-3576 (9th Cir. filed Oct. 24, 2024), Dkt. No. 28.1, pp. 18, 22-24. ↑

37. Id. at p. 36. ↑

38. Cornish-Adebiyi, et al. v. Caesars Entertainment, Inc., et al., No. 1:23-CV-02536-KMW-EAP (D. N.J. Sept. 30, 2024). ↑

39. Daniel Wu, “‘Potato cartel’ mashed competition to raise french fry prices, lawsuits say” (Nov. 20, 2024), https://www.washingtonpost.com/business/2024/11/20/potato-cartel-price-fixing-lawsuit/. ↑

40. Govea, et al. v. National Potato Promotion Board, et al., Case No. 1:24-cv-11816, (N.D. Ill. 2024), https://www.classaction.org/media/govea-v-national-potato-promotion-board-et-al.pdf. Emphasis added. ↑

41. Id. at p. 22. ↑

42. In re Multiplan Health Ins. Provider Litig. (N.D.Ill. June 3, 2025, No. 24 C 6795) 2025 U.S.Dist.LEXIS 104989, at *72. ↑

43. “Report to the California Law Review Commission Antitrust Law: Study B-750: Competition and Artificial Intelligence,” p. 5, https://clrc.ca.gov/pub/Misc-Report/ExRpt-B750-Grp8.pdf. Emphasis in first paragraph added; emphasis in third paragraph in original. ↑

44. Asm. Jud. Comm. Analysis of SB 384 (as amended Jun. 26, 2025), p.9. ↑

Date of Hearing: July 1, 2025

ASSEMBLY COMMITTEE ON JUDICIARY

Ash Kalra, Chair

SBPCA Bill Id:SB 384

Author:(Wahab) – As Amended Ver:June 26, 2025

SENATE VOTE: 28-10

SUBJECT: Preventing Algorithmic Price Fixing Act: prohibition on certain price-setting algorithm uses

KEY ISSUE: SHOULD CALIFORNIA PROHIBIT THE DISTRIBUTION AND USE OF PRICING ALGORITHMS THAT ARE TRAINED ON NONPUBLIC COMPETITOR DATA AND THAT ARE INTENDED TO BE USED BY TWO OR MORE COMPETITORS IN THE SAME MARKET TO SET PRICES?

SYNOPSIS

SB 384 (Wahab) is one of four bills this session targeting algorithmic price coordination—but unlike a price-fixing cartel, the authors have not colluded. SB 52(Perez) and SB 295(Hurtado), which target similar behavior, are independently up for consideration by this Committee today.

SB 384 responds to the growing risk that businesses may rely on shared pricing algorithms to achieve coordinated pricing outcomes that suppress competition—without any express agreement or overt collusion. These tools, often developed by third-party vendors and deployed across firms in the same market, can produce synchronized prices and discourage competition—all without any human agreement, handshake, or email. According to the author, the result is a digital-age version of a price-fixing cartel, but one that may elude traditional antitrust enforcement due to the lack of direct communication or provable agreement. This bill seeks to address this by creating a new statutory cause of action that prohibits selling, licensing, providing or using a price-setting algorithm that is (1) intended for use by two or more competitors in the same market, (2) trained on nonpublic competitor data, (3) knowingly or recklessly used to set prices, supply levels, or rental terms. It creates a civil violation enforceable only by public prosecutors, while preserving a narrow affirmative defense for those who can demonstrate due diligence—such as securing written assurances from vendors about the data used.

Critics contend the bill targets “technology, not conduct.” But SB 384, as recently amended, is focused on a specific misuse of pricing tools: when multiple firms rely on a shared algorithm that processes their own confidential data to align prices. SB 384 does not affect unilateral pricing, tools based solely on public data, or consumer-facing analytics. It simply attempts to draw a clear legal boundary around an increasingly common practice that could undermine price competition in core commercial markets. Similar amendments have been taken in all three price-fixing bills pending today (SB 384, SB 295 and SB 52) to ensure that these measures address the problem of collusion rather than stifling the promise of technology.

This measure is author-sponsored and supported by a broad coalition, including public interest and housing advocates. It is opposed by an industry coalition led by the Chamber of Commerce, among others. In the event of its successful passage out of this Committee, this measure will be heard in the Privacy and Consumer Protection Committee.

SUMMARY: Prohibits the distribution and use of pricing algorithms that are trained on nonpublic competitor data intended to be used by two or more competitors in the same market. Specifically, this bill:

1. Prohibits a person from selling, licensing, providing, or using a price-setting algorithm with the intent that it be used by two or more competitors in the same market if the person knows or should know that the algorithm processes nonpublic input data to set:
   1. Prices or supply levels of a good or a service.
   2. Rent or occupancy levels of rental property.
2. Defines certain key terms, including:
   1. “Nonpublic input data” means data that is confidential, nonpublic, and sensitive information of competitors.
   2. “Price-setting algorithm” means a software, computer system, computer process, algorithmic program, or artificial intelligence that processes nonpublic input data for the purpose of producing a pricing or rental strategy, but does not include a multiple listing service, as provided.
   3. “Competitors” means two or more persons or business entities, including landlords, that offer similar or substitutable goods, services, or real property for lease in the same relevant market to the same or overlapping customer base.
3. Provides an affirmative defense for 1) by demonstrating by clear and convincing evidence that the person exercised reasonable due diligence, including obtaining written assurances from the person selling, licensing, or providing the algorithm that the algorithm does not process nonpublic input data.
4. Provides that for a person who sells, licenses, or provides a price-setting algorithm that violates 1), each user that uses the algorithm constitutes a violation.
5. Provides that for a person who uses the price-setting algorithm that violates 1), each calendar month of use constitutes a separate violation.
6. Provides for public enforcement by the Attorney General or local prosecutors.
7. Permits the recovery of any of the following forms of relief, or a combination of all them:
   1. Injunctive relief;
   2. Restitution; or
   3. Civil penalties of up to $1,000 per violation.
8. Requires the award of attorney’s fees and costs to the Attorney General or the local prosecutor, if they are the prevailing party in the action.

EXISTING LAW:

1. Establishes the Sherman Antitrust Act of 1890 (Sherman Act). (15 U.S.C. Sections 1-7.)
2. Makes illegal, under the Sherman Act, every contract, combination in the form of trust or otherwise, or conspiracy, in restraint of trade or commerce among the states or with foreign nations. (15 U.S.C. Section 1.)
3. Authorizes a state attorney general to bring a civil action in the name of the state in any district court of the United States having jurisdiction over the defendant to secure monetary relief, as provided, for violations of the Sherman Act. (15 U.S.C. Section 15c.)
4. Establishes the Clayton Act. (15 U.S.C. Sections 12-27.)
5. Defines, under the Clayton Act, “antitrust laws” to include the Sherman Act, certain provisions of the Wilson Tariff Act, and the Clayton Act, as amended. (15 U.S.C. Section 12.)
6. Makes illegal, under the Clayton Act, certain exclusive dealing agreements, tying contracts, corporate mergers and acquisitions, and interlocking directorates, as specified. (15 U.S.C. Sections 13-14.)
7. Prohibits every contract by which anyone is restrained from engaging in a lawful profession, trade, or business of any kind is, including specified noncompete clauses, subject to specified exemptions. (Business and Professions Code Section 16600 et seq. All further statutory references are to the Business and Professions Code, unless otherwise specified.)
8. Establishes the Cartwright Act. (Section 16700 et seq.)
9. Defines a “trust” under the Cartwright Act as a combination of capital, skill, or acts by two or more persons for any of the following purposes:
   1. To create or carry out restrictions in trade or commerce.
   2. To limit or reduce the production, or increase the price of, merchandise or of any commodity.
   3. To prevent competition in manufacturing, making, transportation, sale, or purchase of merchandise, produce, or any commodity.
   4. To fix at any standard or figure, whereby its price to the public or consumer shall be in any manner controlled or established, any article or commodity of merchandise, produce, or commerce intended for sale, barter, use, or consumption in the state.
   5. To make or enter into or execute or carry out any contracts, obligations, or agreements of any kind or description, by which they do all or any combination of the following:
      1. Bind themselves not to sell, dispose of, or transport any article or any commodity or any article of trade, use, merchandise, commerce, or consumption below a common standard figure, or fixed value.
      2. Agree in any manner to keep the price of such article, commodity, or transportation at a fixed or graduated figure.
      3. Establish or settle the price of any article, commodity, or transportation between them or themselves and others, so as directly or indirectly to preclude a free and unrestricted competition among themselves, or any purchasers or consumers in the sale or transportation of any such article or commodity.
      4. Agree to pool, combine, or directly or indirectly unite any interests that they may have connected with the sale or transportation of any such article or commodity, that its price in any manner might be affected. (Section 16720.)
10. Makes every trust unlawful, against public policy, and void, except as exempted under the Cartwright Act. (Section 16726.)
11. Establishes the Unfair Practices Act (UPA), which is intended to safeguard the public against the creation or perpetuation of monopolies and to foster and encourage competition, by prohibiting unfair, dishonest, deceptive, destructive, fraudulent and discriminatory practices by which fair and honest competition is destroyed or prevented. (Section 17000 et seq.)
12. Prohibits, under the UPA, a range of behavior that reduces competition in pricing, including specified locality discrimination in pricing, sales under costs or loss leaders made with the intent of injuring competitors or destroying competition, and contracts for the performance of warranty service and repair below the cost of the service or repair. (Sections 17040-17051.)
13. Permits the head of any department in the state to make investigations and prosecute actions concerning violations of law; as part of those investigations, the department head may inspect and copy books, records, and other items, issue subpoenas for the attendance of witnesses and the production of documents or other tangible things, and promulgate interrogatories. (Government Code Sections 11180, 11181.)
14. For purposes of investigating potential violations of the Cartwright Act and the UPA, extends all of the investigative powers granted to the Attorney General pursuant to 13) to the district attorney of any county when the district attorney reasonably believes that a violation has occurred. (Government Code Section 16759 (a).)
15. Establishes that a party has the burden of proof as to each fact the existence or nonexistence of which is essential to the claim for relief or defense that they are asserting. (Evidence Code Section 500.)
16. Establishes the Unfair Competition Law (UCL), which provides a statutory cause of action for any unlawful, unfair, or fraudulent business act or practice and unfair, deceptive, untrue, or misleading advertising, including over the internet. (Business and Professions Code Section 17200 et seq.)
17. Establishes the Consumers Legal Remedies Act (CLRA), which proscribes certain unfair methods of competition and certain unfair or deceptive acts or practices undertaken by a person in a transaction intended to result or that results in the sale or lease of goods or services to a consumer, including advertising goods or services with intent not to sell them as advertised. (Civil Code Section 1770 et seq.)
18. Defines “goods” as “tangible chattels bought or leased for use primarily for personal, family, or household purposes, including certificates or coupons exchangeable for these goods, and including goods that, at the time of the sale or subsequently, are to be so affixed to real property as to become a part of real property, whether or not they are severable from the real property.” (Civil Code Section 1761 (a).)
19. Defines “services” as work, labor, and services for other than a commercial or business use, including services furnished in connection with the sale or repair of goods. (Civil Code Section 1761 (b).)

FISCAL EFFECT: As currently in print this bill is keyed fiscal.

COMMENTS: SB 384 is one of three measures currently pending before the Assembly Committee on the Judiciary seeking to address “algorithmic price fixing.”

Algorithmic price fixing refers to the use of software—often powered by artificial intelligence—to set or recommend prices in ways that result in coordinated outcomes between competitors without any formal agreement. Third-party vendors now offer pricing tools trained on large datasets, sometimes including nonpublic or sensitive market information, which can be deployed across competing firms. These tools can respond rapidly to market changes, discouraging price competition and creating uniform pricing—all while preserving the outward appearance of independent decision-making. The result is a technologically enabled form of price-fixing that achieves the same anticompetitive effects as traditional collusion, but without overt coordination, making it difficult to detect or prosecute under current antitrust law. As explained by the author:

Rapid technological advances have resulted in corporations using tools and resources that result in digital collusion. Corporations rely on the existence of these tools as permission to use them without considering the legal implications. SB 384 places responsibility on these corporations to evaluate the tools they use to process data associated with their own pricing structures and stock by prohibiting the use of tools that collect private data from two or more corporations for the purposes of analyzing and processing the data to create pricing models. This prohibition will force corporations to be more discerning regarding the digital tools they use, and reduce the likelihood of corporations accidentally engaging in digital handshakes that result in price-fixing and market manipulation.

Algorithmic price fixing. Modern algorithmic coordination has its institutional roots in the aftermath of the Airline Deregulation Act of 1978, which eliminated federal oversight of airline fares. Shortly thereafter, major carriers created the Airline Tariff Publishing Company (ATPCO), a centralized platform for submitting and disseminating fare data. Although ATPCO nominally served as a neutral repository, it allowed airlines to signal pricing intentions and monitor rival responses in near real-time, thereby enabling tacit collusion—a form of price coordination not easily captured by conventional antitrust tools. The Department of Justice (DOJ) challenged ATPCO’s structure in the early 1990s under Section 1 of the Sherman Act, arguing that it facilitated price signaling among competitors. The resulting settlement permitted ATPCO to continue with minimal restrictions, effectively sanctioning data-driven pricing coordination in the absence of direct collusion or express agreements.

Today’s version of this phenomenon has been dramatically amplified by advances in artificial intelligence (AI) and data infrastructure. Third-party vendors now offer algorithmic pricing systems trained on vast market datasets—sometimes including nonpublic or competitively sensitive information—which are simultaneously deployed across competing firms. These systems can rapidly adjust prices in reaction to competitors’ changes, discouraging undercutting and reducing market variability, all while maintaining the appearance of independent decision-making. Unlike traditional cartels, these algorithmic tools do not require express communication between competitors and may even operate without any human intervention at all. The result is a technologically mediated form of price fixing that can achieve the same anticompetitive effect as a classic conspiracy—higher prices, suppressed competition, and consumer harm—but in a way that existing legal frameworks struggle to capture. (David Dayen, Three Algorithms in a Room: How Big Data Supercharged Collusion, The American Prospect (June 5, 2024) https://prospect.org/economy/2024-06-05-three-algorithms-in-a-room.) The absence of legislative action in the wake of ATPCO, and the continued judicial reliance on formal notions of “agreement,” have left a legal gray area in which algorithmic collusion remains both highly illegal in principle and difficult to prosecute in practice.

Existing law - The Cartwright Act. California’s main antitrust statute is the Cartwright Act, codified at Business and Professions Code Sections 16700 to 16770. Enacted in 1907, the Act predates the federal Clayton Act and closely parallels Section 1 of the Sherman Act (15 United States Code Section 1), which prohibits contracts, combinations, and conspiracies in restraint of trade. Like its federal counterpart, the Cartwright Act is designed to preserve and promote free competition in commercial markets. However, California courts have construed it as “broader in range and deeper in reach than the Sherman Act.” (Cianci v. Superior Court (1985) 40 Cal. 3d 903, 920.)

Business and Professions Code Section 16720(e)(4) defines a “trust” to include combinations or agreements in which competitors “agree to pool, combine or directly or indirectly unite any interests that they may have connected with the sale or transportation of any such article or commodity, that its price might in any manner be affected.” In other words, the Cartwright Act proscribes even indirect or subtle forms of coordinated behavior that may affect market prices or restrict trade.

To violate the Act, conduct must typically involve:

1. An agreement or concerted action between two or more people (so-called “single firm conduct” is not encompassed by the Act); and
2. Conduct that unreasonably restrains trade.

Under the Cartwright Act, an “agreement” or “concerted action” refers to a mutual understanding or coordinated conduct between two or more persons to restrain trade. It is the foundational element of any antitrust claim under both the Cartwright Act and the federal Sherman Act Section 1. (Marin County Bd. of Realtors, Inc. v. Palsson (1976) 16 Cal.3d 920, 928 (Cartwright Act requires concerted action); Texaco, Inc. v. Dagher (2006) 547 U.S. 1, 5 (federal antitrust law also requires concerted action).)

An agreement under the Cartwright Act does not require a written contract or direct evidence of collusion. California courts recognize that circumstantial evidence may be used to infer concerted action, particularly because direct evidence of collusion is often unavailable in antitrust cases. (In re Cipro Cases I & II (2015) 61 Cal.4th 116, 146.) The concerted action may be established through indirect mechanisms, such as the exchange of competitively sensitive information, market signaling, or the shared use of pricing algorithms that lead to parallel pricing behavior.

However, parallel conduct alone is insufficient to prove concerted action. Courts require additional “plus factors” that support an inference of agreement, such as the existence of facilitating practices, opportunities to collude, or monitoring mechanisms that suggest firms are consciously coordinating their behavior. Once concerted action is established (either through direct or circumstantial evidence), the court turns to the second prong: whether the conduct unreasonably restrains trade.

Under California law, a restraint is considered unreasonable if it harms competition, rather than merely harming individual competitors. (G.H.I.I. v. MTS, Inc. (1983) 147 Cal.App.3d 256, 269.) (“It is often repeated in antitrust cases that the law seeks to protect competition, not competitors; and stiff competition is encouraged, not condemned.”) (internal quotations omitted).) Courts evaluate this through either the per se rule or the rule of reason. Certain types of restraints are deemed per se unlawful—such as horizontal price fixing, market allocation, and bid rigging—because they are presumed to always harm competition and lack any legitimate justification. Other restraints are analyzed under the rule of reason, which requires a case-specific inquiry into the conduct’s actual or likely anticompetitive effects, the parties’ market power, and any procompetitive justifications. (Palsson, supra, 16 Cal.3d at 934-35.)

AB 325 (Aguiar-Curry), which is currently pending in the Senate Judiciary Committee, seeks to amend the Cartwright Act to prohibit the use or distribution of a common pricing algorithm by separate entities if it is part of a conspiracy to restrain trade or if they distribute or use a common pricing algorithm with the intent to fix prices or terms for similar products or services in California, especially if they coerce others to adopt those prices or knowingly participate in a coordinated pricing scheme.

Existing law – the California Legal Remedies Act, the Unfair Competition Law, and the Unfair Practices Act. While the Cartwright Act remains California’s primary vehicle for antitrust enforcement, the state also possesses powerful consumer and commercial protection statutes that play a complementary role in deterring unfair, deceptive, and anticompetitive conduct—including the use of algorithmic pricing tools that distort markets or mislead consumers.

California’s Unfair Competition Law (Business and Professions Code Section 17200 et seq.) prohibits any “unlawful, unfair or fraudulent business act or practice.” The statute is broadly framed to encompass not only conduct that violates other laws (including the Cartwright Act), but also business behavior that is deceptive, anticompetitive, or contrary to public policy, even if not independently unlawful. (See Cel-Tech Communications, Inc. v. Los Angeles Cellular Tel. Co. (1999) 20 Cal.4th 163, 180–182.) Courts have recognized that the UCL provides a flexible enforcement tool for redressing market harms that fall outside traditional antitrust theories, including the coordinated use of information systems or platform tools that suppress price competition or obscure competitive choices. In Cel-Tech, the California Supreme Court affirmed that the UCL is broader than traditional antitrust statutes and can reach “incipient violations of antitrust laws” or conduct that “violates the policy or spirit” of those laws, even if not independently actionable under the Cartwright Act or Sherman Act. (Id. at 187.) The Court emphasized that a “practice may be deemed unfair even if not specifically proscribed by some other law.” (Id. at 180.) While the Court adopted a narrower “tethering” standard for claims between competitors, it made clear that practices that threaten competition through coordination or structural market manipulation may be challenged under the UCL—particularly where the conduct violates antitrust principles in spirit if not in letter. (Id. at 186-87.) Recent court cases are proceeding under that theory. For example, in People v. Amazon.com, Inc. (Super. Ct. San Francisco City & County, No. CGC-22-601202) (2022), the Attorney General alleges that Amazon’s contractual pricing rules and algorithmic pricing enforcement mechanisms operate to suppress price competition across the digital marketplace. The state’s theory is that Amazon discourages sellers from offering lower prices elsewhere by using algorithms to demote or delist those products, thereby distorting competitive outcomes and harming consumers—even without a classic “agreement” among competitors. This case has survived demurrer and is moving forward on the merits.

In tandem, the Consumer Legal Remedies Act (Civil Code Section 1750 et seq.) protects consumers from deceptive methods of competition and unfair or misleading acts or practices in the sale or lease of goods or services. The CLRA prohibits, among other things, misrepresenting the source, characteristics, or benefits of a product or service, as well as bait-and-switch pricing tactics and other forms of consumer manipulation. (Civil Code Section 1770(a).)

What problem is this bill trying to solve? The author explains that SB 384 addresses the challenge posed by algorithmic price coordination—a modern antitrust issue in which competitors rely on shared AI or pricing tools to achieve collusive outcomes without explicit agreements. As firms increasingly turn to artificial intelligence and predictive analytics to make pricing decisions, some third-party platforms or vendors have begun offering tools that rely on pooled data from multiple market participants. Supporters contend that these tools can suppress price competition, stabilize market rates, and limit downward pricing pressure, particularly in concentrated markets such as rental housing or e-commerce. Because existing antitrust statutes like the Cartwright Act and the Sherman Act generally require evidence of agreement or concerted action, current law struggles to reach tacit coordination facilitated through shared algorithms. And the existing consumer protection framework might fall short of properly capturing the conduct at issue. SB 384 aims to close this gap by establishing a new statutory prohibition that applies where algorithmic tools are intentionally used to influence market behavior among competitors through the use of nonpublic input data.

Who does bill target? SB 384 targets both providers and users of price-setting algorithms who contribute to coordinated pricing behavior among competitors through the use of nonpublic, competitively sensitive data. Specifically, the bill applies to any person who sells, licenses, or provides a price-setting algorithm with the intent that it be used by two or more competitors in the same market, and who knows or should know that the algorithm processes nonpublic input data. It likewise applies to any person who uses such an algorithm under the same conditions.

The bill focuses on conduct by horizontal competitors—defined as two or more persons or business entities (including landlords) that offer similar or substitutable goods, services, or real property in the same relevant market to the same or overlapping customer base. The targeted entities include software vendors, technology consultants, developers, or platforms that distribute algorithmic pricing tools trained on confidential competitor data, as well as businesses that implement these tools to set their own prices, supply levels, or rental terms.

What does this bill do? Under SB 384, it is unlawful to sell, license, or otherwise provide a price-setting algorithm to two or more competitors in the same market with the intent that it be used to set prices, supply levels, rent, or occupancy, if the person knows or should know that the algorithm processes nonpublic input data. Similarly, it is unlawful to use such an algorithm for price-setting where the user knows or should know that the tool is trained on competitively sensitive data and has also been used by another competitor for the same purpose. Liability attaches even if there is no express agreement among the parties—effectively replacing the “agreement” element of the Cartwright Act with the functional equivalent of coordinated algorithmic conduct. The bill includes a limited affirmative defense where a person demonstrates that they exercised reasonable due diligence that the algorithm does not process non-public input data. This is meant to protect small businesses who may unknowingly rely on third-party tools without access to their underlying data practices, provided they took reasonable steps—such as obtaining written assurances from the vendor—to ensure the tool did not rely on confidential competitor data.

While the bill draws from the doctrinal structure of California’s Cartwright Act, it creates a distinct liability framework that does not require proof of an explicit agreement or traditional antitrust injury. Instead, it is premised on the concept that the shared use of algorithmic tools trained on nonpublic competitor data can produce the same anticompetitive outcomes as express collusion—namely, higher prices, reduced output, and suppressed price competition—but without direct communication or overt consensus.

By doing so, SB 384 shifts the enforcement paradigm from traditional cartel detection to a per se prohibition on a specific form of algorithmic collusion. The bill eliminates the need to prove market power, define a relevant market, or demonstrate harm to consumers or competition, thereby avoiding the complex economic analysis often required under the rule of reason required under the Cartwright Act. Instead, it identifies the conduct itself—deploying or relying on shared price-setting algorithms trained on nonpublic competitor data—as inherently anticompetitive, and therefore unlawful.

Hypothetical #1 – Lawful under SB 384. A permissible example under the bill would be a travel app such as Hopper or Kayak that aggregates publicly available airfare data from airlines to help consumers identify the best time to purchase tickets. These apps do not rely on nonpublic or competitively sensitive information, nor do they communicate price recommendations back to the airlines. Their use is unilateral and consumer-facing, promoting price transparency rather than coordination. Therefore, such conduct would not be covered by SB 384.

Hypothetical #2 – Unlawful under SB 384. Suppose a dynamic pricing vendor licenses an AI-powered tool to three large grocery chains in the same regional market. The algorithm collects nonpublic pricing data, discounting practices, and volume information from each chain’s internal systems, and uses that data to recommend “market-optimized” pricing strategies across all three. If the vendor knows or reasonably expects that the algorithm will be used to influence pricing decisions across these competitors, and if the grocery chains know or should know the algorithm is trained on confidential data and used across the market, then both the vendor and users could be liable under SB 384.

How will this bill be enforced? SB 384 would be enforced exclusively through civil enforcement actions brought by public prosecutors. Specifically, the Attorney General, district attorneys, city attorneys, or county counsel are authorized to file suit in any court of competent jurisdiction for violations of the bill. These enforcement actions may seek a range of remedies, including:

• Civil penalties of up to $1,000 per violation. For users, each calendar month of use is a separate violation. For persons who sell or license the algorithm, violations are tallied by each user who uses the offending algorithm. So for example, if a person sells a price-setting algorithm that violates SB 384 to 100 separate users, that person would be deemed to have committed 100 separate violations.
• Injunctive relief to prevent ongoing or future violations.
• Restitution (i.e., ill-gotten gains).
• Reasonable attorney’s fees and costs if the public prosecutor prevails in the action.

Importantly, SB 384 does not authorize private rights of action, nor does it duplicate remedies under the Cartwright Act. Instead, it operates as a standalone, per se regulatory enforcement regime focused on deterring and dismantling the technological infrastructure that enables algorithmic collusion.

How does this bill compare to SB 52 and SB 295? SB 384 prohibits the distribution or use of a price-setting algorithm with the intent that it be used by two or more competitors in the same market, where the algorithm processes nonpublic input data. It applies across sectors and explicitly includes rental housing by referencing “rent or occupancy levels.” This makes its scope broader than SB 52, which is limited to residential rental markets, but more targeted than SB 295, which applies to all commercial terms but does not reference housing or real estate.

Unlike SB 295 and SB 52, SB 384 includes a robust affirmative defense: parties who perform reasonable due diligence and obtain written assurances that the algorithm does not process nonpublic data can avoid liability. SB 295, on the other hand, includes a safe harbor for use of data more than one year old but no diligence defense. And SB 52 excludes liability where only public, aggregate, or non-identifiable data is used, and defines multiple exemptions for legitimate sources of rental information.

SB 384 provides for public enforcement by the Attorney General or local prosecutors and imposes civil penalties of up to $1,000 per violation, with violations accruing per user or per month of use. It mirrors SB 52’s penalty amount but lacks SB 52’s private right of action and explicit tenant remedies. SB 295 provides a higher ceiling for penalties—up to $1 million—and allows recovery of treble damages.

Opposition’s concerns. A broad industry coalition, spearheaded by the California Chamber of Commerce, oppose an earlier version of this bill. While some of their concerns may have been addressed by recent amendments, many likely remain.

Opposition argues that “[a]t its core, SB 384 …regulates the technology itself rather than its misuse…Like several other pricing algorithms moving through the Legislature, SB 384 appears rooted in an assumption that pricing algorithms are inherently problematic or unlawful, as opposed to attempting to identify and halt demonstrably anti-competitive behaviors or price-fixing practices.” Contrary to the claim that SB 384 broadly prohibits the use of pricing software or algorithmic tools, the bill, as recently amended, is targeted at a specific and high-risk form of conduct: the intentional use of a shared price-setting algorithm that is trained on nonpublic, competitively sensitive data from multiple competitors in the same market, that the user knows or should know processes nonpublic data. It does not prohibit the use of pricing algorithms as a general matter. Businesses may continue to use their own internal tools, rely on public data, and engage in dynamic pricing based on supply and demand. What SB 384 prohibits is the use or distribution of algorithmic tools that function as digital conduits for competitor coordination—tools that process confidential cross-firm data to generate pricing strategies and thereby erode independent decision-making.

The bill does not regulate the technology itself, but rather its misuse in a collusive context. It requires multiple limiting elements before any liability attaches: (1) the algorithm must be used or distributed with the intent that it be used by two or more competitors in the same market; (2) the person must know or should know that the algorithm processes nonpublic input data; and (3) the data must be competitively sensitive and drawn from rival firms. These elements ensure that SB 384 targets only conduct that replicates the economic effects of price-fixing—a long-recognized antitrust violation—not the use of technology per se.

Importantly, SB 384 does not apply to algorithms trained on public pricing data, nor to tools used unilaterally by a single firm. It would not affect common use cases such as retailers adjusting prices based on market trends, realtors estimating listing values using public sales data, or banks setting terms based on internal risk models. It expressly allows for dynamic pricing and competitive differentiation. What it prohibits is the use of a shared algorithm that encourages parallel pricing behavior across firms by analyzing confidential information that competitors would not lawfully exchange in any other context.

Pending litigation on algorithmic collusion and the California Law Review Commission (CLRC). The simultaneous consideration of SB 295, SB 52, and SB 384 (as well as AB 325 currently pending in the Senate Judiciary Committee) is not happening in a vacuum. As discussed in detail in the Assembly Privacy Committee analysis of AB 325 (Aguiar-Curry, 2025), at pp. 6-8, case law on these issues is evolving and there is an active CLRC working group tasked with studying whether the Cartwright Act needs updating, especially in light of technological changes.

As discussed in Privacy’s analysis of AB 325, a number of high-profile federal and state enforcement actions are actively testing whether algorithmic pricing platforms like RealPage, RENTmaximizer, and Cendyn violate existing antitrust and consumer protection statutes. In the RealPage litigation, for example, the U.S. Department of Justice and multiple state attorneys general, including California’s, are pressing forward with claims under the Sherman Act without needing any new statutory authority. In that case, the DOJ has made clear its position: “Automating an anticompetitive scheme does not make it less anticompetitive.” (Memorandum of Law in Support of Statement of Interest of the United States, In re RealPage, Case No. 3:23-MD-3071 (M.D. Tenn Nov. 15, 2023), https://www.justice.gov/d9/2023-11/418053.pdf.) Similarly, in ongoing cases involving Cendyn’s hotel pricing software, courts are grappling with whether joint use of algorithms constitutes concerted action under existing federal antitrust law.

Furthermore, the CLRC working group has proposed several modest reforms, such as clarifying that concerted action can include the knowing use of common algorithms trained on competitor data, even in the absence of direct communication. The Commission’s final recommendations may offer a more comprehensive and integrated approach to updating California antitrust law.

ARGUMENTS IN SUPPORT: Oakland Privacy, one of the bill’s supporters, explains:

Senate Bill 384 is one of many, with an emphasis on many, bills this year focused on algorithmic pricing formulas and their detrimental effects on affordability for consumers and on market competition. Senate Bill 384 goes further than a similar bill before the committee at this hearing, SB 52, in that it sweeps beyond rental price-setting algorithms like RealPage to encompass any price fixing program as long as that program’s pricing strategy is based on the use of nonpubic data. It is also worthwhile to note that the bill also is more narrow than a surveillance pricing proposal currently in the Assembly AB 446, which seeks to limit differential pricing based on public data that includes personally identifying information including behavorial profiling and tracking.

There is not much doubt that multiple sellers sharing nonpublic data about a good or service for the purpose of deciding on a pricing strategy is classic price-fixing. SB 384 in more or less restating current law that price-fixing is illegal, and seeks to clearly define that price-fixing is still price-fixing if a machine does it and that a sole individual seller can be said to engage in price-fixing if the pricing strategy is determined by an algorithmic process that contains nonpublic data from two or more sellers.

In other words, the absence of two physical different sellers does not legalize the behavior as long as the machine is utilizing the multi-seller non-public competitor data to develop the price-fixing strategy on behalf of a seller.

Senate Bill 384 would eliminate such use of algorithms to price-fix in the above-described manner and prohibit their use in California, although potentially such algorithmic programs could be used or trained using public data to develop pricing strategies, and that use would remain untouched by this bill.

We support this general concept as the adaption of classic antitrust law to the same old price-fixing behavior in new guises and formats due to technological innovation and the increasing capacity of artificial intelligence programming. Consumers are entitled to the benefits of marketplace competition between various sellers, which is the primary lever for downward pressure on prices in pure market-based capitalism. When sellers collude secretively to keep prices artificially high, then consumers lose their strongest weapon which is the ability to walk to another seller should one particular seller raise their price too high.

ARGUMENTS IN OPPOSITION: The Chamber of Commerce, which is leading the opposition coalition, explain some of their concerns with the prior version of the bill:

The fundamental concern we have with SB 384 is that it prohibits the use of certain technology in competitive pricing under the guise of prohibiting price fixing. Under the introduced version of the bill, it very clearly did so by prohibiting any business from using pricing models of any sort to set a price or supply level of a good or service based on any information related to pricing or supply – whether that information was publicly available or nonpublic/confidential. That distinction between publicly available and nonpublic or confidential information is significant, because it preserves activities that businesses long performed in making pricing decisions, and done so legally: observe, analyze, and respond to market conditions; collect information on prices, price changes, and supply levels; analyze/process that information; and create pricing models to inform pricing decisions.

It is worth noting that there are many legitimate grounds for setting different prices for the same goods or services, such as dynamic pricing where prices fluctuate based on real-time demand, availability and market conditions (e.g., peak hours or bad weather can drive up demand for rides); local demand or operational/regional costs; returning customers or those enrolled in loyalty programs may receive lower prices; or lower prices may get set to attract first time customers; online ticket prices may increase as the date of an event gets closer; inventory goes down; etc.). Restricting the ability of businesses to use this type of technology to help them in these same activities will greatly impair the ability of some businesses to understand market conditions and respond efficiently in changes to the competitive landscape, not to mention take away information that would otherwise guide pricing decision and lend to less competitive pricing overall.

REGISTERED SUPPORT / OPPOSITION:

Support

AIDS Healthcare Foundation

California Civil Liberties Advocacy

California Housing Partnership

California Rural Legal Assistance Foundation

California School Employees Association

Fremont for Everyone

Housing California

National Association of Social Workers, California Chapter

Oakland Privacy

Open Markets Institute

Private Equity Stakeholder Project

Sacramento Regional Coalition to End Homelessness

Opposition

American Property Casualty Insurance Association

Association of National Advertisers

Calbroadband

California Apartment Association

California Association of Realtors

California Chamber of Commerce

California Credit Union League

California Fuels and Convenience Alliance

California Hospital Association

California Hotel & Lodging Association

California Restaurant Association

California Retailers Association

California Travel Association

Chamber of Progress

Civil Justice Association of California (CJAC)

Insights Association

National Association of Mutual Insurance Companies

Personal Insurance Federation of California

Realpage, INC.

Software Information Industry Association

Technet

Analysis Prepared by: Shiran Zohar / JUD. / (916) 319-2334

 THIRD READING  

Bill No: SB 384

Author: Wahab (D)

Amended: 4/24/25

Vote: 21

SENATE JUDICIARY COMMITTEE: 11-2, 4/22/25

AYES: Umberg, Allen, Arreguín, Ashby, Caballero, Durazo, Laird, Stern, Wahab, Weber Pierson, Wiener

NOES: Niello, Valladares

SENATE APPROPRIATIONS COMMITTEE: 5-1, 5/23/25

AYES: Caballero, Cabaldon, Grayson, Richardson, Wahab

NOES: Seyarto

NO VOTE RECORDED: Dahle

SUBJECT: Preventing Algorithmic Price Fixing Act: prohibition on price-fixing algorithm use

SOURCE: Author

DIGEST:

This bill prohibits the provision and use of a pricing algorithm that incorporates competitors’ nonpublic, competitively sensitive data, as defined.

ANALYSIS:

Existing law:

1. Establishes the Cartwright Act. (Business and Professions (Bus. & Prof.) Code, div. 7, pt. 2, ch. 2, §§ 16700 et seq.)
2. Defines “person” within the Cartwright Act to include corporations, firms, partnerships, and associations. (Bus. & Prof. Code, § 16702.)
3. Defines a “trust” under the Cartwright Act as a combination of capital, skill, or acts by two or more persons for any specified purposes, including creating or carrying out restrictions in trade or commerce or preventing competition. (Bus. & Prof. Code, § 16720.)
4. Makes every trust unlawful, against public policy, and void, except as exempted under the Cartwright Act. (Bus. & Prof. Code, § 16726.)
5. Authorizes the Attorney General, or the district attorney of any county, subject to specified notice requirements, to initiate a civil action or criminal proceeding for a violation of the Cartwright Act; and authorizes the Attorney General or a person injured by a Cartwright Act violation to file a civil action for monetary relief, as specified. (Bus. & Prof. Code, §§ 16750, 16754, 16760.)
6. Provides that a violation of the Cartwright Act is a conspiracy against trade, and that knowingly engaging or participating in such a conspiracy is a crime, punishable as follows:
   1. If the violator is a corporation, by a fine of not more than $1 million or the amount under (c), whichever is greater.
   2. If the violator is an individual, by imprisonment pursuant to Penal Code section 1170(h) for one, two, or three years; by imprisonment for up to one year in a county jail; by a fine of not more $250,000 or the amount under (c), whichever is greater; or by both a fine and imprisonment.
   3. If any person derives pecuniary gain from a violation of the Cartwright Act, or the violation results in pecuniary loss to a person other than the violator, the violator may be fined not more than twice the amount of the gain or loss. (Bus. & Prof. Code, § 16755(a).)

This bill:

1. Provides that, in addition to any other law, a person shall sell, license, or otherwise provide, and a person shall not use, a price-fixing algorithm to set either of the following:
   1. A price or supply level of a good or service.
   2. A rent or occupancy level of rental property.
2. Provides that it is an affirmative defense to liability under 1) if the seller can show that they conducted due diligence, including inquiring of the person offering the pricing algorithm, and did not learn that the use of the pricing algorithm violated 1).
3. Provides that the following constitutes a separate violation of 1):
   1. For a person who sells, licenses, or otherwise provides a price-fixing algorithm, each authorized user of, or user under a license for, the price-fixing algorithm.
   2. For purposes of a seller who uses a price-fixing algorithm, each calendar month of the use of the price-fixing algorithm.
4. Provides that the Attorney General, in the name of the people of the State of California, or a city attorney or county counsel, in the name of the city or county, may file a civil action for a violation of 1) for actual damages, injunctive relief, restitution, or civil penalties of up to $1,000 per violation, or any combination of those remedies; the court shall also award reasonable attorney’s fees to the Attorney General, city attorney, or county counsel, if they are the prevailing party in the action.
5. Defines the following terms for purposes of 1):
   1. “Artificial intelligence” means an engineered or machine-based system that varies in its level of autonomy and that can, for explicit or implicit objectives, infer from the input it receives how to generate outputs that can influence a physical or virtual environment.
   2. “Nonpublic input data” means data that is competitively sensitive to the seller and that the seller maintains in a confidential manner, but for allowing the data to be used in a price-fixing algorithm, and may include, but is not limited to, price, output, customers, or sales territory.
   3. “Price-fixing algorithm” means a software, computer system, computer process, algorithmic program, or artificial intelligence that both (1) accepts the historical or contemporaneous nonpublic input data of two or more sellers on the price, price change, or supply level of a good or service or rent or occupancy level of a rental property from one or more sellers; and (2) processes that nonpublic input data for the purpose of producing a pricing or rental strategy.
   4. “Seller” means either a person who sells or leases a good or service to a consumer, or a landlord; and includes a person or a business entity that utilizes a price-fixing algorithm on behalf of the seller, lessor, or landlord.

Comments

A human setting prices has to (1) take in new information, (2) analyze the effect of the new information on their own prices, and (3) determine whether to raise or lower prices, and by how much. A pricing algorithm, on the other hand, often uses artificial intelligence and machine learning to weigh variables including supply and demand, a competitor’s prices, and anticipated delivery time, as well as any other factors its programmers have baked into the formula, and can set new prices nearly instantaneously in response to new information. Algorithms that rely on public data alone can still give a seller a massive advantage over a consumer who does not have access to the same volume of data.

Recently, some businesses have started to offer algorithmic pricing models that expressly incorporate competitors’ competitively sensitive data—data which the businesses do not share with the market generally for strategic reasons, and which results in anticompetitive behavior when shared with competitors. While much of this behavior is already prohibited under state and federal antitrust laws, those laws can take years of costly litigation to enforce—while all the while, consumers continue to suffer from higher prices.

This bill is intended to leave no wiggle room by expressly prohibiting the use of price-fixing algorithms that incorporate competitors’ nonpublic, competitively sensitive data. This bill also gives a seller an affirmative defense to liability if the seller conducted due diligence and reasonably determined that the algorithm did not violate this bill’s prohibitions. This bill does not establish a private right of action, but rather limits enforcement to suits brought by the Attorney General, a city attorney, or a county counsel.

This bill is one of several bills pending in the Legislature relating to pricing algorithms, including SB 52 (Pérez, 2025), SB 295 (Hurtado, 2025), and AB 325 (Aguiar-Curry, 2025).

FISCAL EFFECT: Appropriation: No Fiscal Com.: Yes Local: No

According to the Senate Appropriations Committee, this bill’s fiscal impact is as follows:

• State and local prosecutors: The Department of Justice (DOJ) indicates no fiscal impact. However, the department and local prosecutors could incur some potential workload cost pressures (Unfair Competition Law Fund, General Fund, local funds). If state and local prosecutors file civil enforcement actions as authorized by this bill it will result in a significant workload increase. DOJ’s Consumer Protection Section, within the Public Rights Division, could see an increase in workload in investigating and prosecuting of violations of the Unfair Competition Law under this bill and would likely require additional attorneys and legal support staff. (See Staff Comments)
• Trial Courts: Unknown, potentially significant cost to the state funded trial court system (Trial Court Trust Fund, General Fund) to additional adjudicate civil actions. By creating a new a private cause of action with statutory damages, and authorizing a new civil penalty with statutory damages, this bill may lead to additional civil cases. Expanding civil penalties and creating new causes of action could lead to lengthier and more complex court proceedings with attendant workload and resource costs to the court. The fiscal impact of this bill to the courts will depend on many unknown factors, including the number of cases filed and the factors unique to each case. An eight-hour court day costs approximately $10,500 in staff in workload. If court days exceed 10, costs to the trial courts could reach hundreds of thousands of dollars. In 2023–24, over 4.8 million cases were filed statewide in the superior courts. Filings increased over the past year, driven mostly by misdemeanors and infractions, and civil limited cases. The increase in filings from the previous year is greater than 5% for civil limited and unlimited, appellate division appeals, juvenile delinquency, misdemeanors and infractions, and probate. While the courts are not funded on a workload basis, an increase in workload could result in delayed court services and would put pressure on the General Fund to fund additional staff and resources and to increase the amount appropriated to backfill for trial court operations. The Governor’s 2025-26 budget proposes a $40 million ongoing increase in discretionary funding from the General Fund to help pay for increased trial court operation costs beginning in 2025-26.

SUPPORT: (Verified 5/27/25)

AIDS Healthcare Foundation

California Housing Partnership

California Rural Legal Assistance Foundation

California School Employees Association, AFL-CIO

Fremont for Everyone

Housing California

National Association of Social Workers – California Chapter

Oakland Privacy

Private Equity Stakeholder Project

Sacramento Regional Coalition to End Homelessness

OPPOSITION: (Verified 5/27/25)

American Property Casualty Insurance Association

Association of National Advertisers

California Apartment Association

California Association of Realtors

California Chamber of Commerce

California Credit Union League

California Hospital Association

California Retailers Association

Chamber of Progress

Civil Justice Association

Insights Association

National Association of Mutual Insurance Companies

Personal Insurance Federation of California

Software Information Industry Association

TechNet

ARGUMENTS IN SUPPORT: According to the National Association of Workers – California Chapter:

SB 384 is a proactive and essential piece of legislation that addresses the emerging issue of algorithmic price-fixing, which has been shown to result in market manipulation and increased costs for consumers. In recent years, two high-profile cases—Agri Stats Inc. and RealPage—have emerged, where businesses utilized algorithmic tools to engage in anticompetitive practices. These tools facilitated collusion and market manipulation, which drove consumer prices. While current laws may address instances of collusion once the agreements are made, there is a need for proactive intervention to prevent these practices before they occur. SB 384 represents that necessary intervention.

This bill will ensure that businesses take greater responsibility for the software they purchase or lease. It specifically restricts the use of software that combines confidential data from multiple companies. By doing so, SB 384 aims to prevent digital collusion, protect consumers, and foster a fairer and more transparent marketplace. Companies must understand their role in preventing algorithmic price-fixing and act ethically in their software purchases and usage.

ARGUMENTS IN OPPOSITION: According to a coalition of this bill’s opponents:

It is worth noting that there are many legitimate grounds for setting different prices for the same goods or services, such as dynamic pricing where prices fluctuate based on real-time demand, availability and market conditions (e.g., peak hours or bad weather can drive up demand for rides); local demand or operational/regional costs; returning customers or those enrolled in loyalty programs may receive lower prices; or lower prices may get set to attract first time customers; online ticket prices may increase as the date of an event gets closer; inventory goes down; etc.). Restricting the ability of businesses to use this type of technology to help them in these same activities will greatly impair the ability of some businesses to understand market conditions and respond efficiently in changes to the competitive landscape, not to mention take away information that would otherwise guide pricing decision and lend to less competitive pricing overall.

While we appreciate the author’s willingness to narrow the bill by removing reference to public databases and narrowing the bill to nonpublic data of two or more sellers, to truly avoid capturing public data and to avoid the chilling effect that this blanket ban will have on the use of this technology in general, the line between what is public and nonpublic data needs to be more clearly and accurately drawn and actual knowledge of the sellers should be considered, or the end result will remain the same as it was under the introduced version of the bill.

Prepared by: Allison Whitt Meredith / JUD. / (916) 651-4113

5/27/25 16:54:17

**** END ****

 THIRD READING  

Bill No: SB 384

Author: Wahab (D)

Amended: 4/24/25

Vote: 21

SENATE JUDICIARY COMMITTEE: 11-2, 4/22/25

AYES: Umberg, Allen, Arreguín, Ashby, Caballero, Durazo, Laird, Stern, Wahab, Weber Pierson, Wiener

NOES: Niello, Valladares

SENATE APPROPRIATIONS COMMITTEE: 5-1, 5/23/25

AYES: Caballero, Cabaldon, Grayson, Richardson, Wahab

NOES: Seyarto

NO VOTE RECORDED: Dahle

SUBJECT: Preventing Algorithmic Price Fixing Act: prohibition on price-fixing algorithm use

SOURCE: Author

DIGEST:

This bill prohibits the provision and use of a pricing algorithm that incorporates competitors’ nonpublic, competitively sensitive data, as defined.

ANALYSIS:

Existing law:

1. Establishes the Cartwright Act. (Business and Professions (Bus. & Prof.) Code, div. 7, pt. 2, ch. 2, §§ 16700 et seq.)
2. Defines “person” within the Cartwright Act to include corporations, firms, partnerships, and associations. (Bus. & Prof. Code, § 16702.)
3. Defines a “trust” under the Cartwright Act as a combination of capital, skill, or acts by two or more persons for any specified purposes, including creating or carrying out restrictions in trade or commerce or preventing competition. (Bus. & Prof. Code, § 16720.)
4. Makes every trust unlawful, against public policy, and void, except as exempted under the Cartwright Act. (Bus. & Prof. Code, § 16726.)
5. Authorizes the Attorney General, or the district attorney of any county, subject to specified notice requirements, to initiate a civil action or criminal proceeding for a violation of the Cartwright Act; and authorizes the Attorney General or a person injured by a Cartwright Act violation to file a civil action for monetary relief, as specified. (Bus. & Prof. Code, §§ 16750, 16754, 16760.)
6. Provides that a violation of the Cartwright Act is a conspiracy against trade, and that knowingly engaging or participating in such a conspiracy is a crime, punishable as follows:
   1. If the violator is a corporation, by a fine of not more than $1 million or the amount under (c), whichever is greater.
   2. If the violator is an individual, by imprisonment pursuant to Penal Code section 1170(h) for one, two, or three years; by imprisonment for up to one year in a county jail; by a fine of not more $250,000 or the amount under (c), whichever is greater; or by both a fine and imprisonment.
   3. If any person derives pecuniary gain from a violation of the Cartwright Act, or the violation results in pecuniary loss to a person other than the violator, the violator may be fined not more than twice the amount of the gain or loss. (Bus. & Prof. Code, § 16755(a).)

This bill:

1. Provides that, in addition to any other law, a person shall sell, license, or otherwise provide, and a person shall not use, a price-fixing algorithm to set either of the following:
   1. A price or supply level of a good or service.
   2. A rent or occupancy level of rental property.
2. Provides that it is an affirmative defense to liability under 1) if the seller can show that they conducted due diligence, including inquiring of the person offering the pricing algorithm, and did not learn that the use of the pricing algorithm violated 1).
3. Provides that the following constitutes a separate violation of 1):
   1. For a person who sells, licenses, or otherwise provides a price-fixing algorithm, each authorized user of, or user under a license for, the price-fixing algorithm.
   2. For purposes of a seller who uses a price-fixing algorithm, each calendar month of the use of the price-fixing algorithm.
4. Provides that the Attorney General, in the name of the people of the State of California, or a city attorney or county counsel, in the name of the city or county, may file a civil action for a violation of 1) for actual damages, injunctive relief, restitution, or civil penalties of up to $1,000 per violation, or any combination of those remedies; the court shall also award reasonable attorney’s fees to the Attorney General, city attorney, or county counsel, if they are the prevailing party in the action.
5. Defines the following terms for purposes of 1):
   1. “Artificial intelligence” means an engineered or machine-based system that varies in its level of autonomy and that can, for explicit or implicit objectives, infer from the input it receives how to generate outputs that can influence a physical or virtual environment.
   2. “Nonpublic input data” means data that is competitively sensitive to the seller and that the seller maintains in a confidential manner, but for allowing the data to be used in a price-fixing algorithm, and may include, but is not limited to, price, output, customers, or sales territory.
   3. “Price-fixing algorithm” means a software, computer system, computer process, algorithmic program, or artificial intelligence that both (1) accepts the historical or contemporaneous nonpublic input data of two or more sellers on the price, price change, or supply level of a good or service or rent or occupancy level of a rental property from one or more sellers; and (2) processes that nonpublic input data for the purpose of producing a pricing or rental strategy.
   4. “Seller” means either a person who sells or leases a good or service to a consumer, or a landlord; and includes a person or a business entity that utilizes a price-fixing algorithm on behalf of the seller, lessor, or landlord.

Comments

A human setting prices has to (1) take in new information, (2) analyze the effect of the new information on their own prices, and (3) determine whether to raise or lower prices, and by how much. A pricing algorithm, on the other hand, often uses artificial intelligence and machine learning to weigh variables including supply and demand, a competitor’s prices, and anticipated delivery time, as well as any other factors its programmers have baked into the formula, and can set new prices nearly instantaneously in response to new information. Algorithms that rely on public data alone can still give a seller a massive advantage over a consumer who does not have access to the same volume of data.

Recently, some businesses have started to offer algorithmic pricing models that expressly incorporate competitors’ competitively sensitive data—data which the businesses do not share with the market generally for strategic reasons, and which results in anticompetitive behavior when shared with competitors. While much of this behavior is already prohibited under state and federal antitrust laws, those laws can take years of costly litigation to enforce—while all the while, consumers continue to suffer from higher prices.

This bill is intended to leave no wiggle room by expressly prohibiting the use of price-fixing algorithms that incorporate competitors’ nonpublic, competitively sensitive data. This bill also gives a seller an affirmative defense to liability if the seller conducted due diligence and reasonably determined that the algorithm did not violate this bill’s prohibitions. This bill does not establish a private right of action, but rather limits enforcement to suits brought by the Attorney General, a city attorney, or a county counsel.

This bill is one of several bills pending in the Legislature relating to pricing algorithms, including SB 52 (Pérez, 2025), SB 295 (Hurtado, 2025), and AB 325 (Aguiar-Curry, 2025).

FISCAL EFFECT: Appropriation: No Fiscal Com.: Yes Local: No

According to the Senate Appropriations Committee, this bill’s fiscal impact is as follows:

• State and local prosecutors: The Department of Justice (DOJ) indicates no fiscal impact. However, the department and local prosecutors could incur some potential workload cost pressures (Unfair Competition Law Fund, General Fund, local funds). If state and local prosecutors file civil enforcement actions as authorized by this bill it will result in a significant workload increase. DOJ’s Consumer Protection Section, within the Public Rights Division, could see an increase in workload in investigating and prosecuting of violations of the Unfair Competition Law under this bill and would likely require additional attorneys and legal support staff. (See Staff Comments)
• Trial Courts: Unknown, potentially significant cost to the state funded trial court system (Trial Court Trust Fund, General Fund) to additional adjudicate civil actions. By creating a new a private cause of action with statutory damages, and authorizing a new civil penalty with statutory damages, this bill may lead to additional civil cases. Expanding civil penalties and creating new causes of action could lead to lengthier and more complex court proceedings with attendant workload and resource costs to the court. The fiscal impact of this bill to the courts will depend on many unknown factors, including the number of cases filed and the factors unique to each case. An eight-hour court day costs approximately $10,500 in staff in workload. If court days exceed 10, costs to the trial courts could reach hundreds of thousands of dollars. In 2023–24, over 4.8 million cases were filed statewide in the superior courts. Filings increased over the past year, driven mostly by misdemeanors and infractions, and civil limited cases. The increase in filings from the previous year is greater than 5% for civil limited and unlimited, appellate division appeals, juvenile delinquency, misdemeanors and infractions, and probate. While the courts are not funded on a workload basis, an increase in workload could result in delayed court services and would put pressure on the General Fund to fund additional staff and resources and to increase the amount appropriated to backfill for trial court operations. The Governor’s 2025-26 budget proposes a $40 million ongoing increase in discretionary funding from the General Fund to help pay for increased trial court operation costs beginning in 2025-26.

SUPPORT: (Verified 5/24/25)

California Housing Partnership

California Rural Legal Assistance Foundation

California School Employees Association, AFL-CIO

Housing California

National Association of Social Workers – California Chapter

Private Equity Stakeholder Project

Sacramento Regional Coalition to End Homelessness

OPPOSITION: (Verified 5/24/25)

American Property Casualty Insurance Association

Association of National Advertisers

California Apartment Association

California Association of Realtors

California Chamber of Commerce

California Credit Union League

California Hospital Association

California Retailers Association

Chamber of Progress

Civil Justice Association

Insights Association

National Association of Mutual Insurance Companies

Personal Insurance Federation of California

Software Information Industry Association

TechNet

ARGUMENTS IN SUPPORT: According to the National Association of Workers – California Chapter:

SB 384 is a proactive and essential piece of legislation that addresses the emerging issue of algorithmic price-fixing, which has been shown to result in market manipulation and increased costs for consumers. In recent years, two high-profile cases—Agri Stats Inc. and RealPage—have emerged, where businesses utilized algorithmic tools to engage in anticompetitive practices. These tools facilitated collusion and market manipulation, which drove consumer prices. While current laws may address instances of collusion once the agreements are made, there is a need for proactive intervention to prevent these practices before they occur. SB 384 represents that necessary intervention.

This bill will ensure that businesses take greater responsibility for the software they purchase or lease. It specifically restricts the use of software that combines confidential data from multiple companies. By doing so, SB 384 aims to prevent digital collusion, protect consumers, and foster a fairer and more transparent marketplace. Companies must understand their role in preventing algorithmic price-fixing and act ethically in their software purchases and usage.

ARGUMENTS IN OPPOSITION: According to a coalition of this bill’s opponents:

It is worth noting that there are many legitimate grounds for setting different prices for the same goods or services, such as dynamic pricing where prices fluctuate based on real-time demand, availability and market conditions (e.g., peak hours or bad weather can drive up demand for rides); local demand or operational/regional costs; returning customers or those enrolled in loyalty programs may receive lower prices; or lower prices may get set to attract first time customers; online ticket prices may increase as the date of an event gets closer; inventory goes down; etc.). Restricting the ability of businesses to use this type of technology to help them in these same activities will greatly impair the ability of some businesses to understand market conditions and respond efficiently in changes to the competitive landscape, not to mention take away information that would otherwise guide pricing decision and lend to less competitive pricing overall.

While we appreciate the author’s willingness to narrow the bill by removing reference to public databases and narrowing the bill to nonpublic data of two or more sellers, to truly avoid capturing public data and to avoid the chilling effect that this blanket ban will have on the use of this technology in general, the line between what is public and nonpublic data needs to be more clearly and accurately drawn and actual knowledge of the sellers should be considered, or the end result will remain the same as it was under the introduced version of the bill.

Prepared by: Allison Whitt Meredith / JUD. / (916) 651-4113

5/25/25 16:39:49

**** END ****

SENATE COMMITTEE ON APPROPRIATIONS

Senator Anna Caballero, Chair

2025 - 2026 Regular Session

SB 384 (Wahab) - Preventing Algorithmic Price Fixing Act: prohibition on price-fixing algorithm use

Bill Summary:

SB 384 would prohibit price fixing algorithms, as specified.

Fiscal Impact:

• State and local prosecutors: The Department of Justice (DOJ) indicates no fiscal impact. However, the department and local prosecutors could incur some potential workload cost pressures (Unfair Competition Law Fund, General Fund, local funds). If state and local prosecutors file civil enforcement actions as authorized by this bill it will result in a significant workload increase. DOJ’s Consumer Protection Section, within the Public Rights Division, could see an increase in workload in investigating and prosecuting of violations of the Unfair Competition Law under this bill and would likely require additional attorneys and legal support staff. (See Staff Comments)
• Trial Courts: Unknown, potentially significant cost to the state funded trial court system (Trial Court Trust Fund, General Fund) to additional adjudicate civil actions. By creating a new a private cause of action with statutory damages, and authorizing a new civil penalty with statutory damages, this bill may lead to additional civil cases. Expanding civil penalties and creating new causes of action could lead to lengthier and more complex court proceedings with attendant workload and resource costs to the court. The fiscal impact of this bill to the courts will depend on many unknown factors, including the number of cases filed and the factors unique to each case. An eight-hour court day costs approximately $10,500 in staff in workload. If court days exceed 10, costs to the trial courts could reach hundreds of thousands of dollars. In 2023–24, over 4.8 million cases were filed statewide in the superior courts. Filings increased over the past year, driven mostly by misdemeanors and infractions, and civil limited cases. The increase in filings from the previous year is greater than 5% for civil limited and unlimited, appellate division appeals, juvenile delinquency, misdemeanors and infractions, and probate. While the courts are not funded on a workload basis, an increase in workload could result in delayed court services and would put pressure on the General Fund to fund additional staff and resources and to increase the amount appropriated to backfill for trial court operations. The Governor’s 2025-26 budget proposes a $40 million ongoing increase in discretionary funding from the General Fund to help pay for increased trial court operation costs beginning in 2025-26.

Background:

Rental pricing algorithms use extensive market data to predict and recommend profit-maximizing rents. This nonpublic, material, and granular rental data includes, among other information, a landlord’s rental prices from executed leases, lease terms, and future occupancy. Companies like RealPage collect swaths of such data from competing landlords, combine it, and feed it to an algorithm to provide daily, near real time pricing recommendations to landlords. Algorithmic pricing weakens competition because it can facilitate price coordination among landlords who would otherwise be competing. A free market requires that landlords compete on the merits, not coordinate pricing. If price coordination was eliminated, there would be an economically meaningful decrease in price mark-ups for rental units using pricing algorithms.^[1] The costs to renters in 2023 has been estimated at $3.8 billion. ^[2]

Proposed Law:

• Prohibits selling, licensing, or otherwise providing and using a price-fixing algorithm to set either of the following:
  • A price or supply level of a good or service; and/or
  • A rent or occupancy level of rental property.
• Establishes an affirmative defense to liability if the seller can show that they conducted due diligence and did not learn that using the price-fixing algorithm was a violation.
• Establishes distinct and separate violations as follows:
  • For a person who sells, licenses, or otherwise provides a price-fixing algorithm, each authorized user of, or user under a license for, the price-fixing algorithm constitutes a separate violation.
  • For the purposes of a seller who uses a price-fixing algorithm in violation of subdivision, each calendar month of use constitutes a separate violation.
• Authorizes the Attorney General, in the name of the people of the State of California, or a city attorney or county counsel, in the name of the city or county, to file a civil action for a violation to recover actual damages or for injunctive relief, restitution, or civil penalties of up to $1,00) per violation, or any combination of those remedies. The court shall award reasonable attorney’s fees and costs to the Attorney General, city attorney, or county counsel, as applicable, if they are the prevailing party in the action.
• Defines the following:
  • “Artificial intelligence” means an engineered or machine-based system that varies in its level of autonomy and that can, for explicit or implicit objectives, infer from the input it receives how to generate outputs that can influence a physical or virtual environment.
  • “Nonpublic input data” means data that is competitively sensitive to the seller and that the seller maintains in a confidential manner, but for allowing the data to be used in a price-fixing algorithm. “Nonpublic input data” may include, but is not limited to, price, output, customers, or sales territory.
  • “Price-fixing algorithm” means a software, computer system, computer process, algorithmic program, or artificial intelligence that does both of the following:
    • Accepts the historical or contemporaneous nonpublic input data of two or more sellers on the price, price change, or supply level of a good or service or rent or occupancy level of rental property from one or more sellers.
    • Processes the nonpublic input data described in subparagraph (A) for the purpose of producing a pricing or rental strategy.
  • “Seller” means either of the following:
    
    • A person or business entity who sells or leases a good or service to a consumer.
      
    • A landlord.
      
    • A person or a business entity that utilizes a price-fixing algorithm on behalf of a person or business.

Related Legislation:

• AB 316 (Krell) prohibits a defendant that used AI from asserting a defense that the AI autonomously caused the harm to the plaintiff. AB 316 is pending in the Assembly Committee on Privacy and Consumer Protection.
• AB 412 (Bauer Kahan) requires a of a generative AI model to document any copyrighted materials used to train the model. AB 412 is pending in the Senate Judiciary Committee.
• SB 52 (Perez) prohibits a person from selling or using a price-fixing algorithm, to set rental price. SB 52 is pending in this Committee.
• SB 420 (Padilla) regulates high-risk automated decision systems. SB 420 is pending in this Committee.
• SB 468 (Becker) imposes a duty on business that deploy a high-risk AI systems that processes personal information to protect personal information. SB 468 is pending in this Committee.
• AB 1018 (Bauer-Kahan) regulates automated decision systems. AB 1018 is pending in the Assembly Judiciary Committee.

Staff Comments: As noted above, DOJ indicated no significant fiscal impact. DOJ also noted that, while the impact of SB 384 would not pose a significant impact to the DOJ, as numerous bills this session may result in no significant impact to the DOJ, should an aggregate of these bills chapter, the DOJ would submit a workload BCP for additional resources to process the increase to the DOJ workload.

-- END --

1. https://bidenwhitehouse.archives.gov/cea/written-materials/2024/12/17/the-cost-of-anticompetitive-pricing-algorithms-in-rental-housing/#:~:text=In%20this%20CEA%20analysis%2C%20we%20quantify%20the%20anticompetitive,harmed%20when%20competition%20in%20rental%20housing%20is%20weakened.
   ↑

2. https://www.justice.gov/archives/opa/media/1364976/dl?inline ↑

SENATE JUDICIARY COMMITTEE

Senator Thomas Umberg, Chair

2025-2026 Regular Session

Bill No: SB 384 (Author:Wahab)

Version: April 8, 2025

Hearing Date: April 22, 2025

Fiscal: Yes

Urgency: No

Consultant: AWM

SUBJECT

Preventing Algorithmic Price Fixing Act: prohibition on price-fixing algorithm use

DIGEST

This bill prohibits the use of a pricing algorithm that incorporates competitors’ nonpublic, competitively sensitive data, as defined.

EXECUTIVE SUMMARY

A human setting prices has to (1) take in new information, (2) analyze the effect of the new information on their own prices, and (3) determine whether to raise or lower prices, and by how much. A pricing algorithm, on the other hand, often uses artificial intelligence and machine learning to weigh variables including supply and demand, a competitor’s prices, and anticipated delivery time, as well as any other factors its programmers have baked into the formula, and can set new prices nearly instantaneously in response to new information. Algorithms that rely on public data alone can still give a seller a massive advantage over a consumer who does not have access to the same volume of data.

Recently, some businesses have started to offer algorithmic pricing models that expressly incorporate competitors’ competitively sensitive data—data which the businesses do not share with the market generally for strategic reasons, and which results in anticompetitive behavior when shared with competitors. While much of this behavior is already prohibited under state and federal antitrust laws, those laws can take years of costly litigation to enforce—while all the while, consumers continue to suffer from higher prices.

This bill is intended to leave no wiggle room by expressly prohibiting the use of price-fixing algorithms that incorporate competitors’ nonpublic, competitively sensitive data. The bill does not establish a private right of action, but rather limits enforcement to suits brought by the Attorney General, a city attorney, or a county counsel. The author has agreed to a number of amendments in response to opposition, including clarifying the definition of “nonpublic input data” and the scope of damages available.

This bill is sponsored by the author and is supported by the AIDS Healthcare Foundation, the California Housing Partnership, the California Rural Legal Assistance Foundation, Housing California, the National Association of Social Workers – California Chapter, and the Sacramento Regional Coalition to End Homelessness. This bill is opposed by the American Property Casualty Insurance Association, the Association of National Advertisers, the California Chamber of Commerce , the California Credit Union League, the California Hospital Association , the California Retailers Association, Chamber of Progress, the Civil Justice Association, Insights Association, the National Association of Mutual Insurance Companies, the Personal Insurance Federation of California, the Software Information Industry Association, and TechNet.

PROPOSED CHANGES TO THE LAW

Existing law:

1. Establishes the Cartwright Act. (Bus. & Prof. Code, div. 7, pt. 2, ch. 2, §§ 16700 et seq.)
2. Defines “person” within the Cartwright Act to include corporations, firms, partnerships, and associations. (Bus. & Prof. Code, § 16702.)
3. Defines a “trust” under the Cartwright Act as a combination of capital, skill, or acts by two or more persons for any of the following purposes:
   1. To create or carry out restrictions in trade or commerce.
   2. To limit or reduce the production, or increase the price of, merchandise, or of any commodity.
   3. To prevent competition in the manufacturing, making, transportation, sale, or purchase of merchandise, produce, or any commodity.
   4. To fix at any standard or figure, whereby its price to the public or consumer shall be in any manner controlled or established, any article or commodity of merchandise, produce, or commerce intended for sale, barter, use, or consumption in the state.
   5. To make or enter into or execute or carry out any contracts, obligations, or agreements of any kind or description, by which they do all or any combination of the following:
      1. Bind themselves not to sell, dispose of, or transport any article or any commodity or any article of trade, use, merchandise, commerce, or consumption below a common standard figure, or fixed value.
      2. Agree in any manner to keep the price of such article, commodity, or transportation at a fixed or graduated figure.
      3. Establish or settle the price of any article, commodity, or transportation between them or themselves and others, so as directly or indirectly to preclude a free and unrestricted competition among themselves, or any purchasers or consumers in the sale or transportation of any such article or commodity.
   6. Agree to pool, combine, or directly or indirectly unite any interests that they may have connected with the sale or transportation of any such article or commodity, that its price in any manner might be affected. (Bus. & Prof. Code, § 16720.)
4. Makes every trust unlawful, against public policy, and void, except as exempted under the Cartwright Act. (Bus. & Prof. Code, § 16726.)
5. Provides that any contract or agreement in violation of the Cartwright Act is absolutely void and not enforceable. (Bus & Prof. Code, § 16722.)
6. Establishes a general prohibition on unfair competition, known as the Unfair Competition Law (UCL), which covers any unlawful, unfair, or fraudulent business act or practice and unfair, deceptive, untrue, or misleading advertising, and any act prohibited under the False Advertising Law (FAL). (Bus. & Prof. Code, div. 7, pt. 2, ch. 5, §§ 17200.)
7. Provides remedies for a violation of the UCL as follows:
   1. Any person who has been injured by a violation of the UCL may seek restitution and injunctive relief in a court of competent jurisdiction. (Bus. & Prof. Code, §§ 17203, 17204.)
   2. The Attorney General, a district attorney, a county counsel authorized by agreement with the district attorney in actions involving violation of a county ordinance, a city attorney of a city having a population in excess of 750,000, county counsel of any county within which a city has a population in excess of 750,000, a city attorney in a city and county or, with the consent of the district attorney, a city prosecutor in a city having a full-time city prosecutor, may, in an action in the name of the people of the State of California, seek injunctive relief, restitution, and a civil penalty. (Bus & Prof. Code, §§ 17204, 17206.)
   3. Unless otherwise expressly provided, the UCL’s remedies are cumulative to each other and to any other remedies or penalties available under state law. (Bus. & Prof. Code, § 17205.)

This bill:

1. Provides that, in addition to any other law, a seller shall not use a price-fixing algorithm to set either of the following:
   1. A price or supply level of a good or service.
   2. A rent or occupancy level of rental property.
2. Provides that the Attorney General, in the name of the people of the State of California, or a city attorney or county counsel, in the name of the city or county, may file a civil action for a violation of 1) for damages, injunctive relief, restitution, or civil penalties of up to $1,000 per violation, or any combination of those remedies; the court shall also award reasonable attorney’s fees to the Attorney General, city attorney, or county counsel, if they are the prevailing party in the action.
3. Defines the following terms for purposes of 1):
   1. “Artificial intelligence” means a machine-based system that can, for a given set of human-defined objectives, make predictions, recommendations, or decisions influencing real or virtual environments.
   2. “Nonpublic input data” means data that is competitively sensitive, including, but not limited to, price, output, customers, or sales territory.
   3. “Price-fixing algorithm” means a software, system or process, algorithmic program, or artificial intelligence that both (1) accepts the historical or contemporaneous nonpublic input data of two or more sellers on the price, price change, or supply level of a good or service or rent or occupancy level of a rental property from one or more sellers; and (2) processes that nonpublic input data for the purpose of producing a pricing or rental strategy.
   4. “Seller” means either a person who sells or leases a good or service to a consumer, or a landlord; and includes a person or a business entity that utilizes a price-fixing algorithm on behalf of the seller, lessor, or landlord.

COMMENTS

1. Author’s comment

According to the author:

Rapid technological advances have resulted in corporations using tools and resources that result in digital collusion. Corporations rely on the existence of these tools as permission to use them without considering the legal implications. SB 384 places responsibility on these corporations to evaluate the tools they use to process data associated with their own pricing structures and stock by prohibiting the use of tools that collect private data from two or more corporations for the purposes of analyzing and processing the data to create pricing models. This prohibition will force corporations to be more discerning regarding the digital tools they use, and reduce the likelihood of corporations accidentally engaging in digital handshakes that result in price-fixing and market manipulation.

2. Background on price-fixing algorithms

“Pricing algorithms are intended to help firms determine optimal pricing on a near real-time basis.”^[1] A human setting prices has to (1) take in new information, (2) analyze the effect of the new information on their own prices, and (3) determine whether to raise or lower prices, and by how much. A pricing algorithm, on the other hand, often uses artificial intelligence and machine learning “to weigh variables such as supply and demand, competitor pricing, and delivery time,” as well as any other factors its programmers have baked into the formula, and can set new prices nearly instantaneously in response to new information.^[2] Studies indicate that the use of certain pricing algorithms results in higher prices for consumers,^[3] particularly when one seller is using a more sophisticated reinforced learning algorithm and its competitors are using a rule-based algorithm that incorporates that seller’s price as an input.^[4]

As algorithms grew more expansive in the 2010s, scholars raised concerns that algorithms—particularly reinforced learning algorithms—could “learn” to tacitly collude with competitors’ algorithms, thereby keeping prices high.^[5] More recently, however, some businesses have offered algorithmic pricing models that expressly incorporate competitors’ nonpublic, competitively sensitive data—for example, by factoring confidential rental rate and occupancy levels, provided by the users of the algorithm, to make pricing and occupancy recommendations to each landlord on the basis of that confidential information. As explained below, there are allegations that these algorithms not only harm consumers, but also violate existing state and federal antitrust laws.

3. Antitrust law and algorithms

Under the federal Sherman Act,^[6] “[horizontal] price-fixing agreements are unlawful per se.”^[7] This per se bar extends to any “combination formed for the purpose and with the effect of raising, depressing, fixing, pegging, or stabilizing the price of a commodity” in interstate or foreign commerce.^[8] Likewise, under the State’s own antitrust law, the Cartwright Act,^[9] “agreements fixing or tampering with prices are illegal per se.”^[10] These prohibitions “rest on the premise that unrestrained interaction of competitive forces will yield the best allocation of our economic resources, the lowest prices, the highest quality and the greatest material progress, while at the same time providing an environment conducive to the preservation of our democratic political and social institutions.”^[11] The Cartwright Act’s prohibitions on anticompetitive behavior are “broader and deeper in range” than the federal Sherman Act’s.^[12]

Former Federal Trade Commission (FTC) Chairwoman Maureen K. Ohlhausen argues that the use of a vendor that provides algorithmic pricing services derived from confidential pricing information provided by multiple competitors is not, as an antitrust matter, new territory.^[13] She believes such services are merely an updated version of a long-prohibited practice, known as the “hub-and-spoke conspiracy”: “[j]ust as the antitrust laws do not allow competitors to exchange competitively sensitive information directly in an effort to stabilize or control industry pricing, they also prohibit using an intermediary to facilitate the exchange of confidential business information.”^[14] To understand why this is such an easy call, Ohlhausen recommended replacing “algorithm” with “ ‘a guy named Bob’ ”:

Is it ok for a guy named Bob to collect confidential price strategy information from all the participants in a market, and then tell everybody how they should price? If it isn’t ok for a guy named Bob to do it, then it probably isn’t ok for an algorithm to do it either.^[15]

This approach reflects the longstanding antitrust principle that “competitors cannot simply get around antitrust liability by acting through a third-party intermediary or joint venture.”^[16]

Consistent with this interpretation, the U.S Department of Justice (USDOJ) and several states, including California, have filed an antitrust lawsuit against RealPage and several property management companies.^[17] According to the First Amended Complaint, RealPage’s Vice President of Revenue Management Advisory Services described RealPage’s benefit to landlords thusly: “ ‘[T]here is greater good in everybody succeeding versus essentially trying to compete against one another in a way that actually keeps the entire industry down.’ ”^[18] At the time this analysis was released, RealPage’s motion to dismiss was pending before the court.

4. This bill prohibits the use of a pricing algorithm that incorporates competitors’ competitively sensitive information

The bill’s intended scope is narrow: the bill prohibits the use of an algorithm that incorporates competitors’ nonpublic input data, which it defines as “competitively sensitive” information. “Competitively sensitive” is a term of art that refers to the information that a business strategically keeps private in order to maintain a market advantage; prices, outputs, production decisions, and trade secrets can all be competitively sensitive if the business generally keeps that information confidential.^[19] The flipside is that, the more competitively sensitive information is, the more problematic it becomes when that information is shared with competitors.^[20] Consistent with this approach, this bill is not attempting to capture algorithms that scrape publicly available data, but rather only pricing algorithms that combine sellers’ information that they generally keep confidential for competitive reasons— the “guy named Bob” situations. In response to concerns from opposition, the author has agreed to amend the definition of “competitively sensitive” to make clear that algorithms cannot incorporate information that is competitively sensitive to the competitor, and that the list of inputs—price, output, customers, or sales territory—is merely to give examples of data that could be competitively sensitive, if the seller treats it as such.

Enforcement of the bill is limited to civil actions brought by the Attorney General, a city attorney, or county counsel. The public prosecutor in such an action may seek damages, injunctive relief, restitution, or civil penalties of up to $1,000 per violation, or any combination thereof. In response to concerns from opposition, the author has agreed to amend the bill to clarify that only “actual” damages are recoverable, and to define what constitutes a violation for purposes of the civil penalty.

As currently in print, the bill prohibits only a seller from using a pricing algorithm that incorporates nonpublic competitor data; as the opponents point out, the bill does not include a requirement that a seller have actual or constructive knowledge that the pricing algorithm is incorporating nonpublic data before liability can attach. Moreover, prohibiting collusive pricing algorithms from being developed in the first instance is arguably more efficient than addressing the sellers alone. Accordingly, the author has agreed to amend the bill to (1) prohibit a person from offering for sale, selling, or licensing an algorithm that incorporates nonpublic competitor data, and (2) establishing an affirmative defense to liability for a seller who, prior to using an algorithm, conducted due diligence to determine whether the algorithm would incorporate competitors’ competitively sensitive information.

The amendments discussed above are set forth below in Part 6 of this analysis.

5. This bill’s relationship to other laws

This bill is not intended to supplant any existing laws—including, but not limited to, the Cartwright Act or the Unfair Competition Law. To that end, the bill expressly states that its prohibitions are “[i]n addition to any other law.”

The opposition argues that, to the extent this bill covers collusive behavior, it is redundant to existing law. This does not appear to be the case, for a couple of reasons. First, California’s antitrust and unfair competition laws are generally stated in broad terms—barring “trusts” and “unfair practices” rather than behavior at a granular level—which may require substantial litigation before a practice is deemed to be prohibited. This bill, by barring a specific type of pricing algorithm, reduces the ambiguity regarding what conduct is permitted and what conduct is verboten. Second, some of California’s existing competition-related laws require a specific intent to collude, which makes pleading in algorithmic pricing cases difficult. This bill, however, is targeted first and foremost at the algorithms inputs; while the bill does require some culpability on the part of the seller or algorithm developer, it should be able to protect consumers in situations where there is collusion in fact but no specific evidence of an actual plot between sellers.

6. Amendments

As discussed above, the author has agreed to take a number of amendments in to response to concerns from the opposition, and plans to keep working with the opposition as the bill moves through the legislative process. The amendments are set forth below, subject to any nonsubstantive changes the Office of Legislative Counsel may make.

Amendment 1

At page 2, in line 1, after “(a)” insert (1) and after “law,” insert “a person shall not sell, license, or otherwise provide, and a”

Amendment 2

At page 2, in lines 6 and 7, change “(1)” and “(2)” to “(A)” and “(B)”, respectively.

Amendment 3

At page 2, between lines 7 and 8, insert “(2) It shall be an affirmative defense to liability for a seller if the seller can show that they conducted due diligence and did not learn that the price-fixing algorithm would violate paragraph (1), including inquiring of the person selling, licensing, or providing the algorithm whether it was in violation of paragraph (1).”

Amendment 4

At page 8, in line 8, after “(b)” insert “(1)”

Amendment 5

At page 2, in line 11, delete “for” and insert “to recover actual”

Amendment 6

At page 2, between lines 16 and 17, insert:

“(2) (A) For purposes of a person who sells, licenses, or otherwise provides a price-fixing algorithm in violation of paragraph (1), each authorized user of, or user under a license for, the price-fixing algorithm sold, licensed, or provided constitutes a separate violation.

(B) For purposes of a seller who uses a price-fixing algorithm in violation of (1), each calendar month of the use constitutes a separate violation.”

Amendment 7

At page 2, in lines 19 through 22, delete the text in line 19 after “means” and insert “an engineered or machine-based system that varies in its level of autonomy and that can, for explicit or implicit objectives, infer from the input it receives how to generate outputs that can influence physical or virtual environments.”

Amendment 8

At page 2, in lines 24 through 25, delete the text in line 24 after “sensitive” and insert “to the seller and which the seller maintains in a confidential manner but for allowing the data to be used in the pricing algorithm. “Nonpublic input data” may include, but is not limited to, the seller’s prices, outputs, customers, or sales territory.”

Amendment 9

At page 2, in line 26, insert “computer” before “system”

Amendment 10

At page 2, in line 27, insert “computer” before “process”

7. Arguments in support

According to the National Association of Workers – California Chapter:

SB 384 is a proactive and essential piece of legislation that addresses the emerging issue of algorithmic price-fixing, which has been shown to result in market manipulation and increased costs for consumers. In recent years, two high-profile cases—Agri Stats Inc. and RealPage—have emerged, where businesses utilized algorithmic tools to engage in anticompetitive practices. These tools facilitated collusion and market manipulation, which drove consumer prices. While current laws may address instances of collusion once the agreements are made, there is a need for proactive intervention to prevent these practices before they occur. SB 384 represents that necessary intervention.

This bill will ensure that businesses take greater responsibility for the software they purchase or lease. It specifically restricts the use of software that combines confidential data from multiple companies. By doing so, SB 384 aims to prevent digital collusion, protect consumers, and foster a fairer and more transparent marketplace. Companies must understand their role in preventing algorithmic price-fixing and act ethically in their software purchases and usage.

8. Arguments in opposition

According to a coalition of the bill’s opponents:

The fundamental concern we have with SB 384 is that it prohibits the use of certain technology in competitive pricing under the guise of prohibiting price fixing. Under the introduced version of the bill, it very clearly did so by prohibiting any business from using pricing models of any sort to set a price or supply level of a good or service based on any information related to pricing or supply – whether that information was publicly available or nonpublic/confidential. That distinction between publicly available and nonpublic or confidential information is significant, because it preserves activities that businesses long performed in making pricing decisions, and done so legally: observe, analyze, and respond to market conditions; collect information on prices, price changes, and supply levels; analyze/process that information; and create pricing models to inform pricing decisions.

It is worth noting that there are many legitimate grounds for setting different prices for the same goods or services, such as dynamic pricing where prices fluctuate based on real-time demand, availability and market conditions (e.g., peak hours or bad weather can drive up demand for rides); local demand or operational/regional costs; returning customers or those enrolled in loyalty programs may receive lower prices; or lower prices may get set to attract first time customers; online ticket prices may increase as the date of an event gets closer; inventory goes down; etc.). Restricting the ability of businesses to use this type of technology to help them in these same activities will greatly impair the ability of some businesses to understand market conditions and respond efficiently in changes to the competitive landscape, not to mention take away information that would otherwise guide pricing decision and lend to less competitive pricing overall.

While we appreciate the author’s willingness to narrow the bill by removing reference to public databases and narrowing the bill to nonpublic data of two or more sellers, to truly avoid capturing public data and to avoid the chilling effect that this blanket ban will have on the use of this technology in general, the line between what is public and nonpublic data needs to be more clearly and accurately drawn and actual knowledge of the sellers should be considered, or the end result will remain the same as it was under the introduced version of the bill.

SUPPORT

AIDS Healthcare Foundation

California Housing Partnership

California Rural Legal Assistance Foundation

Housing California

National Association of Social Workers – California Chapter

Sacramento Regional Coalition to End Homelessness

OPPOSITION

American Property Casualty Insurance Association

Association of National Advertisers

California Chamber of Commerce

California Credit Union League

California Hospital Association

California Retailers Association

Chamber of Progress

Civil Justice Association

Insights Association

National Association of Mutual Insurance Companies

Personal Insurance Federation of California

Software Information Industry Association

TechNet

RELATED LEGISLATION

Pending legislation:

SB 295 (Hurtado, 2025) establishes the California Preventing Algorithmic Collusion Act of 2025, which prohibits a person from using or distributing any pricing algorithm that uses, incorporates, or was trained with competitor data; requires a person using a pricing algorithm to recommend or set a price or commercial term to make certain commercial disclosures; and requires a person to provide specified information to the Attorney General relating to the use of pricing algorithms. SB 295 is pending before this Committee.

SB 52 (Pérez, 2025) prohibits the sale, licensing, or provision, to two or more persons, a rental pricing algorithm with the intent that it be used by two or more landlords in the same market to set or recommend specified rental terms, and prohibits the use of nonpublic competitor data in an algorithm used to set or recommend specified rental terms. SB 52 is pending before this Committee and is set to be heard on the same date as this bill.

AB 325 (Aguiar-Curry, 2025) expressly prohibits, within the Cartwright Act, the use or distribution of pricing algorithms that use, incorporate, or were trained on nonpublic competitor data. AB 325 is pending before the Assembly Appropriations Committee.

Prior legislation:

SB 1154 (Hurtado, 2024) would have established the California Preventing Algorithmic Collusion Act of 2024, which would have prohibited the use of pricing algorithms to set or recommend a price or commercial term in this state that incorporates nonpublic competitor data, as defined, which may be enforced by the Attorney General; established a partially rebuttable presumption that the use of a prohibited pricing algorithm is a violation of specified state laws prohibiting anticompetitive behavior; and added additional provisions relating to the Attorney General’s investigation of, and disclosures of the use of, pricing algorithms. SB 1154 died in this Committee.

AB 2930 (Bauer-Kahan, 2024) would have prohibited a person or entity from using an automated decision tool, including an algorithm, in a way that results in algorithmic discrimination, as defined, in employment, educational, housing, and other contexts. AB 2930 died on the Assembly Floor.

AB 2230 (Bennett, 2024) would have established the Residential Housing Unfair Practices Act of 2023, which would have amended the Cartwright Act to expressly list certain practices relating to the provision of housing. AB 2230 died in the Assembly Judiciary Committee.

AB 331 (Bauer-Kahan, 2023) was largely similar to AB 2930 (Bauer-Kahan, 2024) and would have prohibited a person or entity from using an automated decision tool, including an algorithm, in a way that results in algorithmic discrimination, as defined, in employment, educational, housing, and other contexts. AB 331 died in the Assembly Appropriations Committee.

AB 2224 (McCarty, 2022) would have required online real estate platforms, known as iBuyers, that use algorithms to determine the value of a property and make offers to purchase a home without the use of a mortgage or other type of financing, to work with a local real estate broker when selling and completing a sale of real property in California. AB 2224 died in the Senate Business, Professions and Economic Development Committee.

**************

1. Bertini & Koenigsberg, The Pitfalls of Pricing Algorithms, Harvard Business Review (Sept.-Oct. 2021), available at https://hbr.org/2021/09/the-pitfalls-of-pricing-algorithms. All links in this analysis are current as of April 17, 2025. ↑

2. Ibid. ↑

3. Calvano, et al., Artificial Intelligence, Algorithmic Pricing, and Collusion (2020) 110 Am. Econ. Rev. 3267. ↑

4. Wang, et al., Algorithms, Artificial Intelligence and Simple Rule-Based Pricing (Jun. 29, 2022) SSRN, p. 40, available at https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4144905. ↑

5. Calvano, supra, at p. 3268. ↑

6. 15 U.S.C. §§ 1-7. ↑

7. U.S. v. Socony-Vacuum Oil Co. (1940 310 U.S. 150, 218. ↑

8. Id. at p. 223. ↑

9. Bus. & Prof. Code, pt. 2, ch. 2, §§ 16700 et seq. ↑

10. Oakland-Alameda County Builders’ Exchange v. F. P. Lathrop Constr. Co. (1971) 4 Cal.3d 354, 363. ↑

11. Marin County Bd. of Realtors, Inc. v. Palsson (1976) 16 Cal.3d 920, 935 (internal quotation marks omitted). ↑

12. In re Cipro Cases I & II (2015) 61 Cal.4th 116, 160 (internal quotation marks omitted). ↑

13. Remarks of Maureen K. Ohlhausen, Acting Chairwoman, U.S. FTC, “Should We Fear The Things That Go Beep In the Night? Some Initial Thoughts on the Intersection of Antitrust Law and Algorithmic Pricing,” Remarks from the Concurrences Antitrust Financial Sector Conference (May 23, 2017), p. 10, available at https://www.ftc.gov/news-events/news/speeches/should-we-fear-things-go-beep-night-some-initial-thoughts-intersection-antitrust-law-algorithmic. ↑

14. Ibid. ↑

15. Ibid. ↑

16. Am. Needle, Inc. v. Nat’l Football League (2010) 560 U.S. 183, 202 (cleaned up). ↑

17. See U.S. v. RealPage, Inc.(M.D.N.C.) Case No. 1:24-cv-00710-LCB-JLW. ↑

18. First Amended Complaint, Dkt. No. 47, U.S. v. RealPage, Inc.(M.D.N.C.) Case No. 1:24-cv-00710-LCB-JLW, ¶ 2 (emphasis in original). ↑

19. E.g., FTC & DOJ, Antitrust Guidelines for Collaborations Among Competitors (Apr. 2000) p. 6. ↑

20. Id. at p. 15; United States Delegation, Note for Roundtable on Information Exchanges Between Competitors Under Competition Law (Oct. 21, 2010) p. 6. ↑