[Editor’s Note: Alex Okuliar is a partner at Morrison Foerster and co-chair of the firm’s Global Antitrust Law Practice Group. Rob Manoso is a trial lawyer and partner in Morrison Foerster’s Global Antitrust Law Practice Group. Tyler Phelps is an associate in the Litigation Department of Morrison Foerster and a member of the firm’s Global Antitrust Law Practice Group. All three are based in Washington, D.C. Any commentary or opinions do not reflect the opinions of Morrison Foerster or LexisNexis®, Mealey Publications™. Copyright © 2025 by Alexander Okuliar and Rob Manoso. Responses are welcome.]
Introduction
Antitrust laws promote competition and prevent unlawful restraints of trade. In recent years, increased reliance on algorithms, machine learning, and other forms of artificial intelligence (AI) has raised new questions about how to apply centuries-old antitrust laws to rapidly changing technology. As courts, regulators, and legislators grapple with how to apply antitrust principles to new technology, early cases have the potential to shape the future of antitrust enforcement.
This article explores how the rise of AI has attracted antitrust scrutiny, with a particular focus on cases and enforcement activity related to pricing algorithms. Several high-profile cases in the hotel and multifamily industries illustrate how courts are addressing allegations involving pricing algorithms and the factors that increase the risk of exposure.
These algorithm-focused cases underscore some of the issues courts will need to evaluate in examining AI more broadly, particularly when it comes to analyzing the benefits of AI tools against any potential anticompetitive effects. Given the ability of AI tools to increase efficiencies and reduce transaction costs, courts should exercise caution before using antitrust to condemn tools that are fundamentally procompetitive or competitively neutral. In many cases, bright line rules may be particularly ill-suited to police the use of AI in rapidly changing markets.
Background: AI and Antitrust Laws
AI can implicate federal antitrust laws (and related state laws1) in numerous circumstances.
Section 1 of the Sherman Act prohibits agreements that unreasonably restrain trade.2 Courts evaluate potential Section 1 violations using two standards: (1) the per se standard for conduct that is inherently anticompetitive (e.g., price-fixing, bid-rigging, and market allocation); and (2) the rule of reason for other conduct, where the key question is whether the conduct’s anticompetitive effects outweigh any procompetitive benefits. Algorithms have been scrutinized as a means to enforce a price-fixing agreement,3 as the object of an illegal price-fixing agreement, or as a way to exchange competitively sensitive information between competitors.
Other antitrust laws also serve as potential sources of scrutiny:
- Section 2 of the Sherman Act, which forbids monopolization and attempted monopolization, may be implicated, if, for example, a company with large market share forces customers to adopt an AI tool or use its tool exclusively, either to further entrench its dominance in an existing market or to gain market share in a new market.
- Section 7 of the Clayton Act prohibits mergers and acquisitions where “the effect of such acquisition may be substantially to lessen competition, or to tend to create a monopoly.”4 Acquisitions may be subject to scrutiny if a transaction will lead to control over a significant AI innovation or AI input.
- Section 5 of the Federal Trade Commission (FTC) Act prohibits certain anticompetitive conduct, including unfair methods of competition. The FTC has used its power under this statute to scrutinize the use of algorithms in “surveillance pricing,” and has identified potential unfair methods of competition related to generative AI, including exclusive dealing and bundling.5
Pricing Algorithms: Increased Antitrust Scrutiny with Mixed Results
The increased reliance on algorithms to inform business decisions has led to numerous investigations and lawsuits against providers and customers of pricing algorithms. Each of these cases center on claims that common use of an algorithm by competitors violates Section 1 of the Sherman Act. Courts have already reached different conclusions about the viability of these algorithm-focused claims, foreshadowing additional divergent rulings in future cases.
Casino Hotels
In Gibson v. Cendyn Grp., LLC and Cornish-Adebiyi v. Caesars Ent., Inc., plaintiffs allege that defendant hotel owners in Las Vegas, Nevada and Atlantic City, New Jersey, respectively, used a common algorithm as part of an anticompetitive agreement. Plaintiffs claim that the defendants engaged in a “hub-and-spoke” conspiracy by sharing pricing information through the algorithm software and relying on the software’s price recommendations to increase room rates.
The district courts dismissed both complaints. As the Gibson court emphasized, “[t]he ‘crucial question’ prompting Section 1 liability is ‘whether the challenged anticompetitive conduct stems from [a lawful] independent decision or from an agreement, tacit or express.’”6 According to the courts, neither complaint answered this question sufficiently based on several common deficiencies:
- The individual defendants in each case began using the revenue management algorithm over the course of many years.7
- Plaintiffs failed to allege that defendants exchanged any competitively sensitive information through the algorithm or that the algorithm’s recommendations were based on competitors’ confidential information.8
- Both complaints failed to allege the delegation of any pricing decisions, given that the algorithms presented pricing recommendations that could be accepted or rejected.9
Plaintiffs in both cases have filed appeals that are pending in the Ninth and Third Circuits, respectively. Notably, the Department of Justice (DOJ) under the Biden administration filed an amicus brief in Gibson advocating for reversal. The DOJ argued that concerted action between competitors in violation of the Sherman Act has the potential to take many forms and need not rely on a formal agreement. The DOJ also claimed that defendants could not avoid liability for collusion just because the algorithms’ output was only a recommendation, and that the presence of discretion could still support a conspiracy claim.11 Oral arguments in Gibson have been set for May 12, 2025.
The RealPage Litigation
Since 2022, numerous lawsuits have been filed against property management software provider RealPage, Inc. and many of the country’s largest multifamily landlords over RealPage’s revenue management software. These cases, including class actions and claims by state and federal enforcers, claim that landlords rely on RealPage’s algorithm—which purportedly incorporates managers’ competitively sensitive information—to make pricing and inventory decisions.
The class actions have been consolidated in the Middle District of Tennessee where, in December 2023, the court ruled on defendants’ motion to dismiss plaintiffs’ claims under both the per se and rule of reason theories. The court declined to apply the per se standard, finding that the alleged conspiracy was “not the straightforward form of horizontal price-fixing conspiracy for which courts apply the per se standard,” based on ambiguity as to when landlords joined the conspiracy, the lack of allegations of communications between landlords, and the absence of absolute pricing delegation to RealPage.12 That said, the court allowed plaintiffs to proceed under the rule of reason, given allegations that defendants had market power in the relevant markets, that defendants had changed their pricing policies, and that RealPage facilitated the exchange of sensitive information among the landlords. Unlike Gibson, the court found that plaintiffs had alleged that RealPage “inputs a melting pot of confidential competitor information through its algorithm and spits out price recommendations based on that private competitor data.”13 That litigation is currently in discovery.
In August 2024, the DOJ and several state attorneys general filed suit in the Middle District of North Carolina against RealPage alleging violations of both Section 1 and 2 of the Sherman Act. The DOJ’s complaint—which was later amended to include claims against several landlords—includes allegations of unlawful information sharing,14 agreements that distort the competitive process,15 and exclusionary conduct.
The parties in that case are currently briefing motions to dismiss. In seeking dismissal, RealPage has argued that the DOJ failed to plausibly allege harm to competition or allege that RealPage has market power in a sufficiently alleged relevant market, and that data collection and RealPage’s model are not exclusionary.16 In response, the DOJ argued that it plausibly alleged harm to competition via both direct and indirect effects, that commercial revenue management software is a relevant market wherein RealPage has monopoly power, and that RealPage harmed the competitive process via its vertical agreements with customers.17
Duffy v. Yardi Systems, Inc.
In Duffy v. Yardi Systems., Inc., the court found that plaintiffs had plausibly alleged the existence of a conspiracy and a per se violation in a case involving another property management software company and several landlords. Specifically, plaintiffs claim the landlords agreed to provide competitively sensitive information to Yardi and use its recommendations to set rental rates, with the shared understanding that Yardi would recommend supracompetitive rates that each landlord would follow.
Notably, the court disagreed with the RealPage court’s finding that per se treatment was inappropriate in the context of a computerized algorithm.18 The court noted, “[t]he fact that the lessor defendants did not meet as a group but rather used an intermediary, Yardi, to compile their commercially sensitive data and calculate the supracompetitive rental rate each participant would utilize does not preclude the existence of an agreement or change its unlawful nature.”19 The case is currently in discovery.
Where Is the Line?
These recent cases illustrate how courts have applied traditional antitrust principles to pricing algorithms, with common themes emerging as to when algorithms are more likely to raise scrutiny. But they also leave many questions unanswered. While there may be circumstances where competitors utilize an algorithm to exchange company-specific current information and agree to follow that algorithm’s prices (the algorithmic “Bob”),20 the specific facts in any given case will be far more complicated. Below we identify some of the issues courts will need to address as they evaluate claims beyond the pleading stage.
Delegating Pricing Decisions
In Gibson, Cornish-Adebiyi, and RealPage, each court hesitated to infer a price-fixing conspiracy, given the extent to which companies retained the ability to accept or overturn the algorithm’s recommendations. Other cases have found that delegation of pricing decisions to an algorithm could support an inference of conspiracy, where companies know that competitors were also following those decisions.21 Courts will need to address when and how acceptance of a recommendation can indicate a conspiracy or otherwise be anticompetitive. If a user only accepts a recommended price 50% of the time, it seems difficult to conclude that the user has given up independent pricing authority. In answering this question, courts will have to dive into the facts surrounding the algorithm’s adoption, acceptance rates, how recommendations are used, and other facts of each customer defendant.
Algorithmic Information Exchanges
Courts will also have to address to what extent the inclusion of competitor information in an algorithm supports a violation. As the RealPage court found, and as DOJ has repeatedly argued, the incorporation of competitively sensitive information into an algorithm increases the risk of anticompetitive effects. But there is a significant difference between competitor-specific information being exchanged through an algorithm “conduit” and an aggregated set of competitor data that is used as one component to generate a recommendation, along with public or internal data. Understanding the extent to which an algorithm incorporates non-public data, and how such data are used in pricing recommendations, will require an intricate analysis of each algorithm in a particular case. This is particularly true given the well-established caselaw that information exchanges can be procompetitive.22
Evaluating the Benefits of Algorithms
The court in the multidistrict RealPage litigation noted, “courts are hesitant to apply the per se standard to new or ‘novel way[s] of doing business’ that have not yet been tested or studied by economists to conclusively determine that these types of conspiracies are per se anticompetitive.”23 Generally, this reasoning, coupled with the fact-intensive inquiry outlined above, is likely to lead most courts away from the use of bright line rules in most circumstances involving the use of algorithmic tools.
Algorithms can provide material procompetitive benefits that overenforcement—including through broad application of per se rules—could undermine.24 For example, pricing and inventory recommendations can cut down on manual decision-making, ensure operational consistency, and generate significant cost-savings. In addition, algorithms can often recommend optimal prices that quickly respond to changes in demand, which can lead to better outcomes for consumers. These benefits may vary by industry and by company and may change over time. Broad rulings could discourage companies from investing in AI-driven efficiencies that benefit consumers, without proof that this investment harms the competitive process.
Beyond Algorithms – Additional Antitrust Scrutiny of AI
While we focus primarily on enforcement activity involving algorithms, courts are increasingly being called upon to answer antitrust questions raised by other AI tools.
Enforcers and litigants have increasingly focused on the extent to which companies may be able to use AI to entrench market positions, expand market power to new markets, or control access to data or other inputs on which AI is dependent. Concerns raised to date include control over large datasets (similar to DOJ’s allegations against RealPage) or computing power that may make it difficult for newer entrants to compete and innovate.25
These issues highlight the extent to which AI will become an increasing feature of future investigations and litigation. Just as many courts have been hesitant to condemn the use of algorithms without a fact-intensive assessment, it is likely that courts and regulators will be cautious when examining other AI-related conduct.
Conclusion
AI has the ability to reshape markets across industries and create new markets and ecosystems. Courts and regulators will continue to attempt to apply traditional antitrust principles as this technology unfolds. Although lawmakers have introduced bills to address perceived gaps,26 new federal laws are likely several years away.27 The Trump administration has not indicated it will reduce its scrutiny of AI or suggested any forthcoming guidance, except to the extent it has signaled a willingness to maintain careful scrutiny of Big Tech companies. In the absence of legislation or clear guidelines, courts will be left to achieve the balance of protecting competition while promoting innovation on a case-by-case basis.
Endnotes
1. States have varying antitrust laws, but the majority mirror federal antitrust laws with prohibitions like those found in the Sherman and Clayton Acts.
2. 15 U.S.C. § 1.
3. See, e.g., United States v. Topkins, No. 15-cr-00201, criminal information filed (N.D. Cal. Apr. 6, 2015).
4. 15 U.S.C. § 18.
5. See Fed. Trade Comm’n, Generative AI Raises Competition Concerns (June 29, 2023), https://www.ftc.gov/policy/advocacy-research/tech-at-ftc/2023/06/generative-ai-raises-competition-concerns.
6. Gibson v. Cendyn Grp., LLC, No. 2:23-CV-00140-MMD-DJA, 2024 U.S. Dist. LEXIS 83547, 2024 WL 2060260, at *3 (D. Nev. May 8, 2024).
7. Id. (“The Court continues to find that the timing of when Hotel Defendants began to use [the revenue management algorithm] renders a tacit agreement among them implausible.”) (internal citations omitted); Cornish-Adebiyi v. Caesars Ent., Inc., No. 1:23-CV-02536-KMW-EAP, 2024 U.S. Dist. LEXIS 178504, 2024 WL 4356188, at *5 (D.N.J. Sept. 30, 2024) (“the fourteen-year gap [between when the first and last defendant began using the algorithm] . . . makes it quite implausible that they tacitly agreed to anything, much less to fix the prices of their hotel rooms.”).
8. Gibson, 2024 U.S. Dist. LEXIS 83547, 2024 WL 2060260, at *5 (“But more to the point, the Court agrees with Defendants that Plaintiffs’ failure to plausibly allege the exchange of confidential information from one of the spokes to the other through the hub’s algorithms is another fatal defect with their first claim because it too compels the conclusion that there is no rim.”); Cornish-Adebiyi, 2024 U.S. Dist. LEXIS 178504, 2024 WL 4356188, at *4–5 (citing to, and agreeing with, the court’s analysis regarding the exchange of information in Gibson).
9. Gibson, 2024 U.S. Dist. LEXIS 83547, 2024 WL 2060260, at *7 (“If [defendants] all agreed to outsource their pricing decisions to a third party, and all agreed to price according to the recommendations provided by that third party, it would be plausible to infer the existence of a collusive agreement to fix prices. But the allegations that could plausibly support that sort of inference do not exist in the [complaint].”); Cornish-Adebiyi, 2024 U.S. Dist. LEXIS 178504, 2024 WL 4356188, at *5 (“the pricing authority the Casino-Hotels continued to retain and exercise, makes it quite implausible that they tacitly agreed to anything, much less to fix the prices of their hotel rooms.”).
10. Cornish-Adebiyi, 2024 U.S. Dist. LEXIS 178504, 2024 WL 4356188, at *7 (internal quotation marks omitted).
11. Brief for the United States as Amicus Curiae in Support of Plaintiffs-Appellants at 23–27, Gibson v. Cendyn Grp., LLC, No. 24-3576 (9th Cir. Oct. 24, 2024).
12. See In re RealPage, Inc., Rental Software Antitrust Litig. (No. II), 709 F. Supp. 3d 478, 519–20 (M.D. Tenn. 2023).
13. Id. at 512.
14. Compl. ¶¶ 225–34, United States v. RealPage Inc., No. 1:24-cv-00710 (M.D.N.C. Aug. 23, 2024).
15. Id. at ¶ 241.
16. See Brief in Support of Defendant RealPage, Inc.’s Motion to Dismiss Under Federal Rule of Civil Procedure 12(b)(6), United States v. RealPage Inc., No. 1:24-cv-00710 (M.D.N.C. Feb. 4, 2025).
17. See Plaintiffs’ Response in Opposition to Defendant RealPage’s Motion to Dismiss, United States v. RealPage Inc., No. 1:24-cv-00710 (M.D.N.C. Feb. 25, 2025).
18. See Duffy v. Yardi Sys., Inc., No. 2:23-CV-01391-RSL, 2024 U.S. Dist. LEXIS 219629, 2024 WL 4980771, at *7–8 (W.D. Wash. Dec. 4, 2024).
19. Id. at *5.
20. See Maureen K. Ohlhausen, Acting Chairman, U.S. Fed. Trade Comm’n, Remarks from the Concurrences Antitrust in the Financial Sector Conference: Should We Fear the Things That Go Beep in the Night? Some Initial Thoughts on the Intersection of Antitrust Law and Algorithmic Pricing (May 23, 2017) (“Just 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 . . . . 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.”), available at https://www.ftc.gov/system/files/documents/public_statements/1220893/ohlhausen_-_concurrences_5-23-17.pdf.
21. See Meyer v. Kalanick, 174 F. Supp. 3d 817 (S.D.N.Y. 2016).
22. See, e.g., United States v. U.S. Gypsum Co., 438 U.S. 422, 441 n. 16 (1978) (“The exchange of price data and other information among competitors does not invariably have anticompetitive effects; indeed such practices can in certain circumstances increase economic efficiency and render markets more, rather than less, competitive.”).
23. In re RealPage, Inc., Rental Software Antitrust Litig. (No. II), 709 F. Supp. 3d 478, 520 (M.D. Tenn. 2023) (internal citations omitted).
24. See Martin Spann et al., Algorithmic Pricing: Implications for Consumers, Managers, and Regulators 27 (IESE Bus. Sch. Working Paper No. 4849019, 2024) (“The potential benefits of pricing algorithms are clear. Among them are simplifying managers’ price decision tasks and empowering managers to adopt more efficient price-setting procedures. They can allow managers and firms to respond more quickly to changes in markets, especially changes in supply and demand, thereby increasing profits…. In addition, prices can be tailored to fine-grained customer segments, even to individual customers in real-time through automated processes. The algorithms can analyze changes in costs, capabilities, and capacities, translating these into changes in supply. They can also assess shifts in consumer behavior and competitive decisions, translating these into changes in demand. Another potential benefit is reducing or even eliminating the human biases that impair managers’ decisions. For example, an algorithm can be designed to avoid the typically pitfalls in human decisions, such as being influenced by sunk costs, driven by regret and loss aversion, and being subject to reference effect and path dependence etc. . . . When carefully tailored, such algorithms should allow for improved coordination between firms and managers by aligning their incentives effectively.”), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4849019#.
25. Jonathan Kanter, Assistant Attorney General for the Antitrust Division, U.S. Dep’t of Justice, Assistant Attorney General Jonathan Kanter Delivers Remarks at the Promoting Competition in Artificial Intelligence Workshop (“AI relies on massive amounts of data and computing power, which can give already-dominant firms a substantial advantage. Powerful network effects may enable dominant firms to control these new markets, and existing power in the digital economy may create a powerful incentive to control emerging innovations that will not only impact our economy, but the health and well-being of our society and free expression.”), https://www.justice.gov/archives/opa/speech/assistant-attorney-general-jonathan-kanter-delivers-remarks-promoting-competition.
26. See News Release, Amy Klobuchar, Klobuchar, Colleagues Introduce Antitrust Legislation to Take on Algorithmic Price Fixing, Bring Down Costs (Feb. 6, 2025) ([Senators] introduced the Preventing Algorithmic Collusion Act to prevent companies from using algorithms to collude to set higher prices.”), https://www.klobuchar.senate.gov/public/index.cfm/2025/2/klobuchar-colleagues-introduce-antitrust-legislation-to-take-on-algorithmic-price-fixing-bring-down-costs.
27. However, certain localities, such as San Francisco and Philadelphia, have taken proactive steps to address algorithm-driven pricing in the rental market by introducing legislation that bans the use of algorithmic software to set rent prices.