Algorithms that set prices for vending machines, apartment rents, and hotel rooms are drawing federal antitrust scrutiny as regulators argue these systems can function like old-fashioned cartels, even when no humans shake hands on a deal. A string of lawsuits and settlements now connects the dots between a vending machine operator accused of quietly overcharging card users, a rental software company that allegedly coordinated pricing for millions of apartments, and a Las Vegas hotel pricing vendor that supposedly helped resorts align their nightly rates. The common thread is a simple and uncomfortable idea: when competitors feed their data into the same algorithm, the machine can do the price-fixing for them.
Vending Machines That Charged More Than the Sticker Price
The case that most directly fits the headline involves Canteen, a vending services brand operated by Compass Group USA. According to the court-authorized settlement site for Jilek v. Compass Group USA, Inc. d/b/a Canteen in the U.S. District Court for the Western District of North Carolina (Charlotte Division), certain Canteen vending machines charged card purchasers more than the displayed price without disclosure. The alleged overcharges affected what the case calls “Subject Vending Machines,” and the class period stretches from 2014 to July 9, 2025. That is more than a decade during which consumers swiping a debit or credit card may have paid a hidden premium above the price shown on the machine’s label.
This is not, strictly speaking, an AI cartel story in the way the headline implies. The Canteen allegations center on undisclosed surcharges rather than coordinated algorithmic pricing across competitors. But the case matters because it shows how automated payment systems in low-dollar, high-volume retail environments can extract extra money from consumers without their knowledge. If a vending machine can quietly add cents to every transaction for years, the infrastructure already exists for more sophisticated coordination. The question regulators are now confronting across multiple industries is whether shared software platforms turn that theoretical risk into a practical one, especially when multiple firms plug their data into the same pricing engine.
RealPage and the Blueprint for Algorithmic Price Coordination
The most developed federal case against algorithmic pricing involves RealPage, a software company that served landlords across the country. The Justice Department filed a civil antitrust complaint alleging that RealPage facilitated algorithmic pricing coordination by collecting competitively sensitive data from rival landlords and feeding it into a system that generated pricing recommendations. The alleged mechanisms included sharing nonpublic data among competitors, issuing pricing recommendations that pushed rents upward, and limiting concessions and discounting. Prosecutors say the scheme harmed millions of American renters by making it easier for large property owners to move in lockstep instead of competing aggressively on price.
The DOJ has since reached a proposed resolution that requires RealPage to stop sharing competitively sensitive information and aligning pricing among competitors. The remedy terms are specific and revealing: the settlement prohibits use of competitors’ nonpublic data at runtime, restricts how recent training data can be, limits geographic granularity, and subjects the company to monitoring. Each of those provisions targets a distinct channel through which an algorithm could effectively replicate cartel behavior. Restricting data recency, for instance, means the software cannot use fresh competitor pricing to adjust recommendations in near-real time, which is the digital equivalent of banning phone calls between rival executives before they set tomorrow’s prices. As federal prosecutors have framed it, the case is a test of whether antitrust law can reach a modern form of collusion in which a shared algorithm replaces the smoke-filled room.
Federal Legal Theory: Fixing the Starting Point Is Enough
What makes the government’s position especially aggressive is the legal argument it has advanced in parallel cases. In a Statement of Interest filed jointly by the DOJ and FTC in the case of McKenna Duffy v. Yardi Systems, Inc. in the Western District of Washington, the agencies argued that price-fixing allegations are not defeated merely because firms retain discretion to deviate from algorithmic recommendations. The filing emphasizes the illegality of agreements fixing the “starting point,” such as list prices, and states that effectiveness and adherence are not required for per se treatment under antitrust law. In plain terms, even if a landlord or hotel operator ignores the algorithm’s suggestion half the time, the act of agreeing to let a shared system set the baseline price could still be illegal.
This legal framework matters far beyond rental housing. If regulators can establish that fixing a starting point through shared software counts as per se price-fixing, the same logic applies to any industry where competitors subscribe to a common pricing platform. Vending machine operators, gas stations, parking garages, and ride-hailing services all use algorithmic tools that could, in theory, ingest competitor data and nudge prices upward. The DOJ and FTC have also weighed in on the Cornish-Adebiyi litigation, publishing a joint Statement of Interest that reinforces the same principles in the context of hotel pricing on the Las Vegas Strip. Across these filings, the agencies are signaling that they view shared algorithms as potential hubs of collusion, even in the absence of explicit, human-to-human agreements on final prices.
Vegas Hotels and the Hub-and-Spoke Model
The hotel industry provides perhaps the clearest illustration of how a single vendor can serve as the hub of an alleged price-fixing wheel. In a lawsuit captioned Gibson v. MGM Resorts et al., plaintiffs allege that major Las Vegas resorts shared data with a vendor known as Rainmaker, later called Cendyn. According to the complaint, the vendor’s software ingested nonpublic booking and pricing information from competing hotels and then issued recommendations that tended to raise room rates across the Strip. Plaintiffs argue that by feeding their data into the same system and following its suggestions, rival resorts could move together on price without ever sitting down to hash out an explicit agreement, creating what antitrust lawyers describe as a “hub-and-spoke” conspiracy.
Regulators’ interest in the Vegas cases is not just about one tourist market. It is about the template. If a single vendor can aggregate competitors’ confidential data, analyze demand, and then push out harmonized pricing advice, the line between smart revenue management and unlawful coordination becomes thin. The DOJ and FTC’s joint filing in the related Cornish-Adebiyi case stresses that antitrust law does not require proof that every hotel followed every recommendation, only that there was an agreement to use a shared mechanism for setting prices. That view, if accepted by courts, would make it much harder for companies to defend algorithmic coordination as mere “advice” and could chill the use of certain revenue-management tools in other concentrated markets.
Cartels Without Meetings and the Future of Algorithmic Pricing
Taken together, the vending machine settlement, the RealPage litigation, and the Vegas hotel lawsuits illustrate how antitrust enforcement is adapting to a world where software, not executives, often sets the numbers customers see. In the Canteen matter, the alleged harm came from undisclosed surcharges on individual transactions, showing how automated systems can quietly extract value from consumers at scale. In the RealPage and hotel cases, the focus shifts to how shared algorithms can standardize pricing strategies across competitors, potentially raising market-wide prices without any overt collusion. The common concern is that when firms outsource pricing to a common vendor, traditional safeguards against cartel behavior may no longer function.
For businesses, the emerging message from regulators is that algorithmic tools are not a legal shield. Companies that rely on third-party software to manage prices may need to scrutinize what data those tools ingest, how recommendations are generated, and whether competitors are effectively coordinating through a shared platform. For consumers and tenants, these cases highlight why prices in everyday markets, from snack machines to studio apartments to weekend hotel stays, can feel strangely synchronized. As courts weigh the government’s theory that fixing a starting point through a common algorithm is enough to trigger per se antitrust liability, the outcome will help determine whether algorithmic cartels remain a cautionary metaphor or become a central target of modern competition policy.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
