AI Pricing Tools Are Facing New Legal Scrutiny in Canada and the U.S.

AI pricing tools are no longer just about boosting efficiency. In both Canada and the United States, they’ve become a serious legal and regulatory concern in rental housing, as lawmakers and competition authorities ask whether software that recommends prices can also facilitate coordination among competitors or obscure how prices are determined. The key risk for businesses isn’t that algorithmic pricing is inherently unlawful, but that

In both markets, the core issue is not whether algorithms can optimize revenue, but whether they can do so by pooling sensitive data and nudging landlords toward parallel pricing behavior that looks a lot like collusion. There is also the risk that the way these tools are trained, shared and disclosed could run afoul of antitrust, consumer-protection, privacy and transparency laws.

Why this issue matters

In property management, pricing software has been sold as a way to keep vacancy low, respond faster to demand, and improve revenue management. But when a tool uses nonpublic competitor data to recommend rents, concessions, or lease terms, regulators worry it can weaken independent decision-making and reduce true price competition. That concern has moved from theory to enforcement, especially in multifamily housing where even small rent changes affect millions of households.

The new scrutiny reflects a broader shift in competition law: agencies are no longer focused only on human-to-human agreements, but also on software-enabled coordination. In the U.S. The Justice Department said RealPage’s algorithmic pricing scheme allegedly let landlords share competitively sensitive information and align apartment rents, while Canada’s Competition Bureau concluded a parallel investigation and issued guidance for landlords, property managers, and software providers.

U.S. enforcement pressure

The U.S. case is the clearest signal that AI pricing tools in real estate can trigger major antitrust exposure. In August 2024, the Department of Justice and several state attorneys general sued RealPage, alleging violations of Sections 1 and 2 of the Sherman Act based on software that allegedly decreased competition among landlords and helped monopolize revenue management software for apartment pricing. The DOJ said the software used nonpublic, competitively sensitive rental information from competing landlords to generate pricing recommendations.

The DOJ’s language was especially important because it framed algorithmic pricing as a modern delivery system for old antitrust harms, not a legal loophole. The department stated that using software does not immunize a scheme from Sherman Act liability and emphasized that “training a machine to break the law is still breaking the law”. That line matters because it suggests regulators may treat algorithmic coordination as illegal even if no one literally texts a competitor and agrees on a rent number.

The broader litigation environment also intensified after the original DOJ filing. In January 2025, federal prosecutors sued six major U.S. landlords, accusing them of using RealPage pricing software to collude and raise rents, extending the government’s theory from software vendor conduct to customer conduct as well. By late 2025, the DOJ announced a settlement that would restrict how RealPage could use nonpublic data in pricing recommendations, including limits on real-time data use. That settlement direction suggests the government is willing to demand structural changes, not just penalties.

Canada’s regulatory path

Canada has followed a similar but somewhat more measured path. The Competition Bureau confirmed in 2025 that it was investigating whether landlords were using AI-driven algorithmic pricing in rental housing markets. Later, on November 10, 2025, the Bureau said it had concluded its civil investigation and found that revenue management tools had not been used widely enough by landlords to substantially harm competition under the Competition Act.

That conclusion did not amount to a clean bill of health. The Bureau also said it remained concerned about the potential impact of algorithmic pricing tools on competition in multifamily rental housing and would continue monitoring the sector closely. It issued guidance for landlords, property managers, and software providers, which is a meaningful sign that Canada sees the issue as an ongoing compliance problem rather than a one-off enforcement story.

Canada’s public materials also make the legal theory explicit. The Bureau’s discussion paper on algorithmic pricing says that using common pricing algorithms or pooling data among competitors may raise issues under the Competition Act and can facilitate price-fixing, which is prohibited and may lead to significant penalties, including fines and imprisonment. That is a strong warning for property managers using shared platforms that compile market-wide lease data and produce rent recommendations.

What regulators fear

The legal concern is not that software is “smart”; it is that software can create a channel for coordination that is harder to detect than a traditional cartel. In apartment markets, a pricing engine may ingest competitor data, identify a neighborhood’s rate floor, and recommend simultaneous increases across landlords who otherwise might undercut one another. If many landlords rely on the same model, competition may become more uniform even without an express agreement.

Regulators also worry about “auto-accept” features and human monitoring layers that make algorithmic recommendations sticky. The DOJ alleged that RealPage used tools and pricing advisors to encourage landlords to follow recommendations rather than compete independently. This matters because antitrust law is sensitive not just to price recommendations, but to mechanisms that reduce deviation from those recommendations.

There is also a consumer-protection angle. When rent recommendations are generated from opaque systems, tenants may have no idea whether a price reflects true market forces or a software-driven consensus. Canadian and U.S. enforcers have both signaled that AI pricing can create competition, consumer-harm, and privacy risks at the same time.

Practical risks for landlords

For owners and property managers in Canada and the U.S., the practical risk is that adopting AI pricing tools can create exposure even if the software vendor is the primary target. The U.S. litigation shows that landlords themselves can be pulled into enforcement if they share data or follow recommendations in a coordinated way. In Canada, the Bureau’s guidance to landlords and property managers suggests that compliance obligations are not limited to technology companies.

The highest-risk behaviors are fairly clear. These include sharing nonpublic leasing data with competitors through a common platform, using tools that recommend rents based on rival properties’ confidential information, and allowing software to function as a de facto channel for market alignment. Systems that merely benchmark against public data and support independent pricing decisions are less legally risky, but the line can blur quickly depending on the inputs and governance.

Property managers should also think about recordkeeping and vendor due diligence. If a pricing product cannot explain what data it uses, whether it includes competitor inputs, and how often recommendations are updated, that opacity becomes a business risk. In an enforcement setting, opaque design can make it harder to prove that the tool was used independently and lawfully.

Canada vs. U.S.

The two countries are moving in the same direction, but with different styles of enforcement. The U.S. has taken a more aggressive litigation-first approach, with a DOJ lawsuit, follow-on actions against landlords, and a settlement framework that appears to limit how algorithmic pricing can operate. Canada has so far emphasized investigation, consultation, guidance, and monitoring, while stopping short of finding a competition violation on the facts it reviewed.

That difference should not be misread as a safe harbor in Canada. The Bureau’s conclusion rested on adoption levels and the evidence available at the time, not on a declaration that algorithmic pricing is inherently lawful. In other words, a tool may escape liability in one factual period and still become problematic if adoption rises, data pooling expands, or market structure changes.

For national property management firms, this means a single AI pricing strategy may face different enforcement timing but similar legal theories across borders. A tool that is marketed as revenue optimization in one market may be viewed as an antitrust risk in another if it relies on competitor-sensitive data or produces market-wide rent alignment.

Compliance steps

Firms in the Canadian and U.S. rental markets should treat AI pricing as a legal control issue, not just a software purchase. The safest starting point is to map exactly what data each tool uses, who can access it, and whether any nonpublic competitor information enters the model. If a vendor cannot explain that clearly, the product deserves extra scrutiny.

A second step is to separate decision support from decision automation. Pricing tools should inform managers, not replace independent judgment, and teams should document when they accept, reject, or modify recommendations. That helps show the company is acting independently rather than mechanically following a market signal produced by a shared system.

A third step is vendor contracting. Property managers should negotiate for representations about lawful data sources, audit rights, notice of model changes, and the ability to turn off sensitive features such as auto-accept or competitor pooling. That will not eliminate antitrust risk, but it helps demonstrate a compliance program built around lawful use rather than blind adoption.

Market outlook

The most likely outcome is not a ban on AI pricing in real estate, but a narrower and more disciplined market. Tools that rely on public data, preserve independent pricing choices, and avoid competitor data pooling are likely to remain usable, while systems that blur the line between optimization and coordination will face continuing pressure. Expect more disclosure, more audits, and more litigation over how the models work.

For the property management industry, this is a turning point. The story is no longer “Can AI price apartments?” The better question is “Under what conditions does AI pricing become unlawful coordination?” In both Canada and the U.S., regulators are answering that question by focusing on data provenance, market structure, and whether software is helping competitors compete less.

The bottom line for landlords, operators, and vendors is simple: AI pricing can be a legitimate efficiency tool, but once it starts pooling sensitive competitor data or steering market-wide rents, it enters antitrust danger territory. That is why AI pricing tools are facing new legal scrutiny now, and why that scrutiny is likely to intensify rather than fade.