Key Takeaways<\/p>\n
\tTo date, many competition authorities around the world have conducted market studies and produced guidance and research papers addressing the risks of algorithmic- or AI-based pricing.
\n\tEnforcement activity is expected to increase, with the United Kingdom\u2019s Competition and Markets Authority (CMA) opening an investigation into hotels using data analytics tools and the European Commission confirming that a number of confidential investigations relating to algorithmic pricing that have raised \u201cred flags\u201d are underway.
\n\tAs the use of such pricing tools is becoming more common, businesses should use caution and ensure that they understand the methods and underlying data sources used by the tools; as noted by the CMA, businesses are responsible for the actions of AI tools in the same way they would be responsible for actions of employees.<\/p>\n
An October 2025 Organisation for Economic Co-operation and Development (OECD) report<\/a> on algorithmic pricing and competition in G7 jurisdictions noted that, despite their potential efficiency-enhancing and procompetitive effects, pricing algorithms can raise issues in areas including competition, consumer protection, and data protection. The report concluded that, despite the varying frameworks, there are commonalities in G7 competition authorities\u2019 perspectives on pricing algorithms; many authorities have adopted similar strategies, which has resulted in the prioritization of market studies, consultations, and guidance, with relatively limited enforcement activity to date. Only a limited number of cases have crystallised, which primarily concern traditional theories of harm.<\/p>\n Authorities\u2019 main concerns about algorithmic pricing are (1) competitors using pricing algorithms to implement, facilitate, and\/or monitor \u201ctraditional\u201d anticompetitive agreements; (2) hub-and-spoke agreements and the exchange of competitively sensitive information through algorithmic tools; and (3) tacit collusion and\/or parallel conduct that reduces the competitiveness of a market.<\/p>\n In the United States, as noted in our earlier Update<\/a>, several states have introduced legislation restricting the use of algorithmic pricing. The DOJ also recently settled a matter in which it had alleged claims regarding the exchange of competitively sensitive information through revenue management software.<\/p>\n Competition and Consumer Law Concerns in the UK<\/p>\n The topic has been on the agenda in the UK for a while\u2014the Competition and Markets Authority published an economic research paper<\/a> on pricing algorithms in 2018 and a report<\/a> on competition and algorithms in 2021, with the first CMA case involving collusion via an online pricing tool concluding in 2016<\/a>. The CMA also launched a dynamic pricing project<\/a> considering how dynamic pricing is used across different areas of the economy in late 2024, with tips for businesses published in June 2025.<\/p>\n More recently, in late February 2026, the CMA launched an investigation into the suspected exchange of competitively sensitive information between competing hotel chains using a hotel data analytics tool. The three hotel chains and the provider of the analytics tool are all currently under investigation<\/a>, with the CMA currently gathering information and further case updates expected in the second half of 2026.<\/p>\n In early March 2026, the CMA published a blog post<\/a> on algorithms, AI, and collusion. The post notes the potential benefits of algorithmic pricing\u2014including improved efficiency and lower costs for businesses, the ability to provide personalized offers, and increased responsiveness to changes in demand and supply. It also addresses the risks associated with dynamic pricing (primarily consumer protection-related) and algorithmic collusion (whether via an explicit agreement, as in the 2016 Posters case; via hub-and-spoke arrangements, which may be what the CMA suspects in the Hotels case; or via the use of predictable or autonomous agents, which may achieve collusive outcomes without human intent or involvement). The CMA notes that competition issues resulting from this type of coordination may be addressed through market investigations, rather than Chapter I enforcement. The CMA reminds businesses that risks can be mitigated primarily by not sharing competitively sensitive information with competitors (directly or indirectly); not letting competitors\u2019 confidential information influence pricing decisions; and scrutinizing the data and algorithms used in pricing tools, including by exercising caution when using the same algorithms or pricing tools as competitors.<\/p>\n In parallel, the CMA is actively considering the risks and impact of new tools, such as AI agents and dynamic pricing, from the consumer protection angle. Notably, it launched a consumer protection case<\/a> against Ticketmaster in September 2024 concerning opaque and potentially misleading pricing for the Oasis Live \u201925 concert tour. In September 2025, the CMA secured undertakings from Ticketmaster to improve transparency and fairness in relation to its pricing.<\/p>\n This work on pricing algorithms sits alongside the CMA\u2019s broader work on AI and competition and consumer protection risks. In March 2026, the CMA published a research paper and guidance for businesses on agentic AI and consumer law, both of which make clear that businesses are responsible for the actions and decisions of their AI agents in the same way they are for those of an employee.<\/p>\n Increasing Scrutiny in the EU\u2014More To Come?<\/p>\n The topic is increasingly a focus of attention in the European Union, as well. The European Commission announced that it has launched a number of investigations<\/a> into algorithmic pricing, although the details of these have not yet been made public. In her remarks to the 2026 ABA Antitrust Spring Meeting, acting Director General of DG Competition Linsey McCallum noted<\/a> that \u201c[o]n pricing algorithms, we have a large-scale exercise underway currently that has allowed us to pick up what look to be some red flags, maybe some anomalous pricing contact. So, some very live issues for the time being.\u201d She went on to remind her audience that EU competition law does not require colluding parties to reach \u201cagreement.\u201d Rather, it is sufficient to show that sensitive information was exchanged or that there were \u201cconcerted\u201d practices. Finally, she noted that algorithms that \u201cfacilitate\u201d contacts between competitors can also fall within the scope of EU law. This signals a clear shift from a purely theoretical discussion about the risks of such practices to enforcement.<\/p>\n At member-state level, the Italian Competition Authority recently conducted a market investigation<\/a> into pricing algorithms in passenger air transport on certain routes and raised potential concerns about the transparency and comparability of display to users of ticket and ancillary service prices. In parallel, the Dutch Competition Authority conducted a market investigation<\/a> into computer-controlled pricing in the Dutch aviation sector, seeking to better understand how prices are set in practice.<\/p>\n Practical Tips for Businesses<\/p>\n In both the EU and the UK, businesses are responsible for pricing (and other) decisions implemented by algorithms and AI agents, under both competition and consumer law. As a result, businesses should exercise extra care when using such tools to make strategic decisions on issues such as pricing and inventory levels. It is essential that businesses understand how these tools operate and know what information recommendations or decisions produced by the tools are based on. Competitively sensitive information\u2014such as forward-looking information on business strategy, pricing, and inventory levels\u2014should generally not be shared with such tools unless specific safeguards are in place, ensuring that such information remains entirely internal, and a legal risk assessment has been carried out. Because scrutiny of such algorithms and AI-based tools is only likely to increase over the next few years around the globe, we recommend training key stakeholders and legal teams \u201cearly and often\u201d and seeking legal advice in the event that there is any suggestion of a concern.<\/p>\n