Introduction

AI-enabled pricing refers to the use of artificial intelligence, machine learning, automated decision-making, and predictive analytics to determine the prices of goods and services. Businesses use these technologies to respond to demand, forecast sales, optimize inventory, and adjust prices in real time.

Although AI pricing can improve efficiency and competition, it may also create risks of price coordination. Competing firms can use similar algorithms, share commercially sensitive data, or employ a common pricing platform that reduces competitive uncertainty. In some situations, algorithms may even learn pricing strategies that sustain higher prices without an explicit agreement between the businesses.

The central competition-law issue is whether the pricing conduct constitutes an agreement or concerted practice that restricts competition, or whether it is independent commercial behavior that remains lawful.

1. Meaning of AI-enabled pricing

AI-enabled pricing is the application of algorithmic systems to determine, recommend, or implement prices.

Examples include:

Dynamic pricing for airline tickets and hotel rooms.

Automated pricing on e-commerce marketplaces.

Surge pricing in ride-hailing services.

Algorithmic repricing by online retailers.

AI-based pricing of electricity, insurance, and financial services.

Automated discounting and promotional offers.

The algorithm may use historical sales, competitor prices, demand forecasts, inventory, customer behavior, and other information to determine a price.

Difference between ordinary dynamic pricing and collusive pricing

The fact that prices change automatically or become similar does not, by itself, establish a cartel.

2. Legal framework under Indian competition law

Section 3 of the Competition Act, 2002

Section 3 prohibits agreements that cause or are likely to cause an appreciable adverse effect on competition (AAEC).

Section 3(3) specifically addresses agreements between enterprises engaged in identical or similar trade that involve:

Direct or indirect price fixing.

Limiting production, supply, or markets.

Market allocation.

Bid rigging or collusive bidding.

AI pricing may become unlawful where it is used to implement or facilitate such agreements.

Section 4: Abuse of dominant position

A dominant enterprise using AI pricing may also raise issues under Section 4, including unfair or discriminatory conditions, exclusionary conduct, or leveraging market power.

Dominance alone is not prohibited. The relevant question is whether the conduct amounts to abuse under the Act.

Section 19: Investigation

The CCI may investigate suspected anti-competitive conduct under Section 19. AI systems can assist in identifying suspicious price patterns, but the legal investigation must establish the relevant elements of a contravention.

Section 26: Investigation procedure

The Director General conducts investigations in accordance with the Competition Act. Algorithmic evidence may support an investigation, but the parties must have the opportunity to respond to the evidence and the legal allegations.

3. How AI-enabled pricing can create collusion risks

Four routes to algorithmic coordination

1. Human agreement

Competitors agree to fix prices and use AI to implement the arrangement.

2. Common platform

A pricing provider or marketplace facilitates coordination between competing businesses.

3. Sensitive information

Competitors share current or future prices, costs, or strategic plans through an algorithmic system.

4. Autonomous learning

Independent algorithms learn to respond to one another in ways that sustain supra-competitive prices.

3.1 Explicit algorithmic collusion

This occurs when competitors deliberately agree to use AI or automated pricing systems to maintain a common price.

Example: Three online retailers agree to use a common algorithm that prevents their prices from falling below a specified minimum. The algorithm is merely the instrument used to implement the underlying price-fixing agreement.

3.2 Tacit coordination

Tacit coordination occurs when firms independently recognize that aggressive price competition is less profitable and adopt similar pricing strategies.

AI may make this easier by enabling firms to observe competitor prices, respond quickly, and learn from market reactions. However, the legal distinction between tacit coordination and a prohibited concerted practice remains important.

3.3 Hub-and-spoke coordination

A common intermediary may act as a hub connecting competing businesses. If the hub facilitates agreement or coordination among the spokes, the arrangement may violate competition law.

A common pricing algorithm can create risks if businesses knowingly use it to align prices or reduce competitive uncertainty.

3.4 Price signaling

Algorithms can also facilitate price signaling. One firm may announce a future price increase through an automated system, and other firms may respond by adopting similar prices.

Public price announcements are not automatically unlawful, but coordinated signaling can become problematic where it facilitates an anti-competitive agreement or concerted practice.

4. At least 6 important case laws

The following decisions establish principles relevant to AI-enabled pricing and collusion. Some directly involve algorithmic pricing, while others concern traditional price fixing, information exchange, or digital platforms. The distinction is important: an algorithmic pricing system can be legally relevant even when the precedent itself predates modern AI.

Case 1: United States v. Topkins (2015)

Court: United States District Court for the Northern District of California.

Subject: Algorithm-assisted price fixing.

Facts: Daniel William Topkins, an online poster seller, pleaded guilty to participating in a conspiracy with other online sellers to fix prices. The participants used pricing algorithms to implement their agreement.

Legal principle: The use of an automated pricing program does not make a price-fixing agreement lawful. Traditional antitrust principles apply even when technology is used to carry out the conspiracy.

Relevance to AI-enabled pricing: This is a foundational example of algorithm-assisted collusion. If competing firms agree to use AI to maintain prices, the underlying agreement may violate Section 3 of the Indian Competition Act, 2002. The algorithm is not a defense to cartel liability.

Case 2: Eturas UAB v. Lietuvos Respublikos konkurencijos taryba (2016)

Court: Court of Justice of the European Union.

Subject: Online travel platform and concerted practices.

Facts: Eturas operated an online travel booking system. The platform sent a message to participating travel agencies restricting the maximum discount they could offer through the system. The case concerned whether the agencies could be held responsible for participating in a concerted practice.

Legal principle: An undertaking may be held responsible for participating in a concerted practice where it was aware of a communication facilitating coordination and continued participating, subject to the applicable evidentiary requirements.

Relevance: This case is particularly useful for understanding common algorithmic pricing platforms. If an intermediary communicates a pricing restriction to competing businesses, the legal inquiry may extend to the participating firms' knowledge and conduct. Merely using a common platform does not automatically establish a cartel.

Case 3: United States v. Airline Tariff Publishing Co. (1994)

Court: United States District Court for the District of Columbia.

Subject: Computerized fare systems and price coordination.

Facts: The case involved allegations concerning the use of computerized airline fare publication systems to facilitate communication and coordination of prices.

Legal principle: The exchange of commercially sensitive pricing information through electronic systems may facilitate anti-competitive coordination.

Relevance: Modern AI pricing systems rely on extensive information about market conditions. This case demonstrates why competition authorities should investigate whether the exchange of current or future prices reduces competitive uncertainty or facilitates coordinated pricing.

Case 4: T-Mobile Netherlands BV v. Raad van bestuur van de Nederlandse Mededingingsautoriteit (2009)

Court: Court of Justice of the European Union.

Subject: Exchange of information and concerted practices.

Facts: Mobile telephone operators participated in a meeting involving the exchange of information relevant to competition.

Legal principle: The exchange of commercially sensitive information between competitors may constitute a concerted practice when it reduces uncertainty about future market conduct.

Relevance: AI-enabled pricing can amplify the effect of information exchange. If competing firms provide their future pricing strategies or commercially sensitive data to a shared system, the arrangement may raise competition-law concerns even if the firms do not sign a formal cartel agreement.

Case 5: Excel Crop Care Ltd. v. Competition Commission of India (2017)

Court: Supreme Court of India.

Subject: Cartel conduct and bid rigging.

Facts: Excel Crop Care and other manufacturers were found to have engaged in anti-competitive conduct in the market for aluminium phosphide tablets used in grain storage.

Legal principle: The Supreme Court upheld findings of cartel conduct and examined the application of penalty provisions under the Competition Act.

Relevance: This is a major Indian cartel precedent for AI-enabled pricing. Automated tender systems can identify repeated bid rotation, suspiciously similar prices, and patterns of coordinated bidding. But the legal finding must be supported by evidence of cartel conduct, not merely algorithmic similarity.

Case 6: Competition Commission of India v. Steel Authority of India Ltd. (2010)

Court: Supreme Court of India.

Subject: CCI jurisdiction and investigation powers.

Facts: The case concerned the jurisdiction and procedural powers of the CCI in dealing with competition-law complaints.

Legal principle: The Supreme Court recognized the statutory role of the CCI in investigating and addressing anti-competitive conduct within the framework of the Competition Act.

Relevance: AI-generated pricing alerts may provide leads for CCI investigations. However, the authority to investigate comes from the Competition Act, and algorithmic output cannot replace the statutory process or the parties' procedural rights.

Case 7: United States v. Apple Inc. (E-books) (2013)

Court: United States District Court for the Southern District of New York.

Subject: Facilitated price coordination.

Facts: Apple was found to have participated in a conspiracy involving publishers to raise e-book prices.

Legal principle: An intermediary can play a role in facilitating price coordination between competitors. Contractual and technological arrangements do not automatically make such coordination lawful.

Relevance: AI pricing providers and online marketplaces may act as intermediaries in a pricing ecosystem. If a platform knowingly facilitates coordination among competing businesses, the arrangement may be examined under competition law.

Case 8: United Brands Company v. Commission of the European Communities (1978)

Court: Court of Justice of the European Communities.

Subject: Abuse of dominance and market power.

Facts: United Brands was found to have abused a dominant position in the banana market through commercial practices affecting distributors and market access.

Legal principle: The case established important principles concerning relevant market definition and abuse of dominance.

Relevance: A dominant enterprise using AI-enabled pricing may raise issues under Section 4 of the Competition Act, 2002. Algorithmic pricing is not automatically abusive, but its effects on rivals, customers, and market access may be relevant where dominance exists.

5. Analysis of AI-enabled pricing under Section 3

5.1 Agreement to fix prices

The clearest violation occurs when competing firms agree on prices and use AI to implement the arrangement.

For example, a group of manufacturers may agree that their AI systems will maintain a minimum selling price. This is a conventional cartel with automated implementation.

5.2 Agreement to share pricing information

Competitors may use a common platform to exchange future prices, inventory data, or strategic plans. Such an arrangement can raise concerns where the information exchange facilitates coordination.

The legal assessment depends on the nature of the information, the market structure, and the actual conduct of the parties.

5.3 Independent algorithmic pricing

Independent pricing decisions are not automatically unlawful. A firm may use AI to optimize prices based on demand, costs, and publicly observable market information.

The key issue is whether the firm has engaged in an agreement or concerted practice prohibited by competition law.

5.4 Autonomous algorithmic coordination

Autonomous coordination raises a difficult legal question: if algorithms independently learn to maintain high prices, without any human agreement, can the conduct be prohibited?

Under the Indian Competition Act, Section 3 generally requires an agreement or arrangement, including an understanding or concerted action within the statutory framework. Autonomous coordination without such a legal basis may not fit easily into traditional cartel provisions.

However, if the algorithms are used by a dominant enterprise to exclude rivals, Section 4 may become relevant.

6. AI-enabled pricing and abuse of dominance

Section 4 of the Competition Act, 2002 prohibits abuse of dominant position.

AI pricing can raise concerns in the following ways:

Dominance must be established in the relevant market, and the conduct must fall within the statutory definition of abuse.

7. Evidentiary value of AI-generated pricing data

AI-generated pricing data may be used to identify suspicious patterns, but it must be interpreted carefully.

Types of evidence

Pricing histories and synchronized price movements.

Algorithm source code and configuration.

Contracts with pricing software providers.

Emails, messages, and internal documents.

Tender bids and procurement records.

Evidence of communication between competitors.

Standard of proof

The legal standard depends on the applicable proceeding and the relevant statutory framework. The CCI and courts must consider the evidence as a whole. Statistical patterns may support an inference, but they do not automatically prove an agreement.

8. Challenges in AI-enabled pricing regulation

A. Lack of transparency

AI models may be difficult to explain. A firm may not understand precisely why its pricing system selected a particular price, creating challenges for regulators and courts.

B. False positives

Similar prices may result from common costs, demand shocks, or independent business decisions. AI may mistakenly flag lawful competition as collusion.

C. Attribution of responsibility

It may be difficult to determine whether liability should attach to the business using the AI system, the software developer, or an intermediary. The legal analysis must focus on the actual conduct and the role of each undertaking.

D. Privacy and commercial confidentiality

AI pricing systems may process sensitive business information and personal data. Investigations must respect applicable privacy and confidentiality requirements.

E. Autonomous learning

Autonomous systems may adapt to competitor behavior without explicit instructions to coordinate. This creates uncertainty about how existing agreement-based competition law applies to purely algorithmic outcomes.

9. Preventive compliance measures for businesses

Businesses using AI-enabled pricing should adopt safeguards to reduce collusion risks.

AI pricing compliance checklist

Conduct competition-law review of pricing algorithms.

Avoid agreements with competitors to fix or maintain prices.

Restrict access to competitors' confidential pricing information.

Document independent commercial reasons for price changes.

Monitor software providers and common pricing platforms.

Train pricing teams on cartel and information-exchange risks.

0 of 6 completed

Reset

10. Critical evaluation

AI-enabled pricing can increase efficiency and improve market responsiveness. It may reduce waste, optimize supply, and offer consumers more competitive prices.

However, AI can also reduce uncertainty between competitors, facilitate information exchange, and sustain supra-competitive prices. The risk is particularly significant in concentrated markets, digital marketplaces, and sectors where firms rely on common pricing software.

Competition law must therefore preserve the distinction between legitimate independent pricing and unlawful coordination. The use of AI should not create an automatic presumption of cartel conduct, but it should also not be treated as a technological exemption from competition law.

Conclusion

AI-enabled pricing and collusion risks represent an important modern competition-law challenge. The use of algorithms does not change the fundamental principles of cartel law: agreements to fix prices, allocate markets, or restrict competition may be prohibited under Section 3 of the Competition Act, 2002.

The cases of United States v. Topkins, Eturas, Airline Tariff Publishing, T-Mobile Netherlands, Excel Crop Care, and Apple demonstrate the legal significance of algorithm-assisted pricing, information exchange, and intermediaries.

The most important principle is that AI may be used to optimize prices, but it cannot lawfully be used to implement or facilitate anti-competitive agreements. Future competition enforcement will require a combination of economic analysis, algorithmic expertise, and careful application of legal standards.