Competition Law and Legal Intelligence Market Power

1. Introduction

Legal intelligence refers to technology-enabled services that collect, organise, analyse, and generate insights from legal information. It includes:

• legal research databases;
• case-law and legislation platforms;
• litigation analytics;
• judicial-outcome and judge analytics;
• legal search engines;
• AI-powered legal research assistants;
• contract and regulatory intelligence systems;
• legal citation and knowledge graphs;
• legal datasets used to train AI models.

Competition concerns arise when a legal-intelligence provider obtains or maintains market power over legal information, datasets, search infrastructure, distribution channels, or AI-enabled legal research.

The central competition-law question is not whether a company is large merely because it possesses extensive legal information. The issue is whether its control over important inputs, users, technology, data, standards, or distribution enables it to exclude competitors, exploit customers, raise switching costs, restrict interoperability, or entrench its position.

2. Relevant Competition-Law Framework

A legal-intelligence market can potentially involve:

1. Market definition
2. Dominance or substantial market power
3. Exclusive access to legal datasets
4. Refusal to supply
5. Self-preferencing
6. Tying and bundling
7. Predatory pricing
8. Excessive pricing
9. Discriminatory access
10. Interoperability restrictions
11. Data foreclosure
12. Acquisitions of emerging AI competitors
13. Vertical foreclosure
14. Algorithmic coordination
15. Network effects and switching costs

In India, the principal statutory framework is the Competition Act, 2002, particularly Sections 3 and 4, together with the merger-control provisions. Other jurisdictions apply comparable concepts through Article 102 TFEU, the Sherman Act, Clayton Act, and their national competition statutes.

3. Relevant Market in Legal Intelligence

Market definition is particularly difficult because legal-intelligence services may overlap with several neighbouring markets.

Possible relevant product markets include:

A. Legal research databases

Platforms providing searchable:

• judgments;
• statutes;
• regulations;
• commentaries;
• legal opinions;
• citations.

B. Litigation analytics

Services providing:

• judge analytics;
• case outcome statistics;
• litigation trends;
• damages analysis;
• counsel analytics.

C. AI legal research

Platforms using large language models to:

• retrieve authorities;
• summarise judgments;
• generate research memoranda;
• identify relevant precedents;
• answer legal questions.

D. Regulatory intelligence

Platforms monitoring:

• regulatory amendments;
• enforcement actions;
• compliance requirements;
• sector-specific regulations.

E. Legal data infrastructure

This may constitute a separate upstream market involving:

• digitised judgments;
• structured case metadata;
• citation databases;
• legal ontologies;
• proprietary legal datasets.

The fact that these services are technologically related does not necessarily mean they constitute one relevant market.

4. Sources of Legal-Intelligence Market Power

4.1 Proprietary legal databases

A company may possess a large historical database containing:

• millions of judgments;
• historical versions of statutes;
• headnotes;
• citations;
• legal commentary;
• litigation outcomes.

The database can become a significant competitive advantage if rivals cannot economically reproduce it.

However, copyright or database ownership does not automatically establish competition-law dominance.

The competition question is whether control of the resource creates substantial foreclosure effects.

4.2 Network Effects

Legal-intelligence platforms can benefit from network effects.

More users can generate:

• more searches;
• more feedback;
• more usage data;
• better search relevance;
• improved AI models;
• stronger citation networks.

This can create a feedback loop:

More users → more data → better product → more users → greater market power.

5. Switching Costs

Law firms and corporate legal departments may become heavily dependent on a particular platform.

Switching may require:

• retraining lawyers;
• migrating research histories;
• changing APIs;
• reconstructing internal knowledge repositories;
• modifying workflows;
• losing saved searches;
• losing customised taxonomies.

High switching costs can therefore strengthen incumbent market power even when nominal subscription prices remain competitive.

6. Data as a Competitive Advantage

Legal-intelligence businesses often possess valuable datasets relating to:

• judicial decisions;
• citation relationships;
• litigation histories;
• legal terminology;
• regulatory developments;
• user search patterns.

The competitive concern becomes particularly significant where the incumbent uses data generated by its dominant platform to improve an AI system while restricting competing systems' access to comparable inputs.

A competition authority would need to distinguish between:

legitimate innovation using proprietary data

and

strategic data foreclosure designed to exclude rivals.

7. Refusal to Supply Legal Data

Suppose a dominant legal-information provider controls an indispensable dataset and refuses access to competitors.

Competition-law analysis may examine:

1. whether the dataset is genuinely indispensable;
2. whether duplication is realistically possible;
3. whether access is technically feasible;
4. whether the refusal eliminates effective competition;
5. whether there is a legitimate business justification;
6. whether access can be provided without disproportionate harm;
7. whether the provider previously supplied the information.

The essential-facilities doctrine may become relevant in exceptional circumstances.

8. Self-Preferencing

A legal-intelligence platform might operate both:

• a legal research marketplace/search service; and
• its own AI research product.

It could theoretically manipulate search results so that its AI-generated answers appear above competing legal-research services.

Potential concerns include:

• preferential ranking;
• suppression of competing sources;
• preferential API access;
• discriminatory indexing;
• preferential inclusion in recommendation systems.

The competition analysis depends heavily on the platform's market position and the actual effects of the conduct.

9. Tying and Bundling

A dominant legal-information provider might condition access to its case-law database upon purchasing:

• AI research tools;
• contract-management software;
• legal practice-management software;
• compliance products.

Competition authorities could examine whether:

Product A + Product B

is being used to leverage dominance from one market into another.

Important questions include whether:

• the products are distinct;
• customers are forced or strongly induced to purchase both;
• the undertaking is dominant in the tying market;
• competitors are foreclosed;
• efficiencies justify the arrangement.

10. AI and Legal-Intelligence Market Power

Generative AI creates several new competition issues.

Potential sources of power

A legal-AI provider may control:

• proprietary legal datasets;
• model-training data;
• computational infrastructure;
• specialised legal models;
• distribution platforms;
• lawyer-user networks;
• citation databases;
• APIs.

This produces a potential data–model–distribution feedback loop.

For example:

Proprietary legal database → superior AI model → more lawyers → more usage data → further model improvement → stronger market position.

Competition authorities may therefore need to examine not merely current market share but also dynamic barriers to entry.

11. Six Important Case Laws

1. United Brands v Commission

Case 27/76, United Brands Company v Commission, EU, 1978

Principle

The Court of Justice examined dominance through factors including:

• market position;
• economic strength;
• barriers to entry;
• competitive constraints;
• customer dependence.

Relevance to legal intelligence

A legal-intelligence provider with a very large database and strong customer dependence could potentially possess substantial market power where competing databases impose insufficient competitive pressure.

The case demonstrates that dominance is not determined solely by market share.

12. IMS Health v Commission

Case C-418/01, IMS Health GmbH & Co. OHG v NDC Health GmbH & Co. KG, EU, 2004

Principle

IMS Health concerned access to a protected information structure and the circumstances in which refusal to license an intellectual-property-related resource may raise Article 102 concerns.

The Court developed stringent conditions concerning exceptional circumstances surrounding refusal to supply.

Relevance

This is particularly important for legal intelligence because a proprietary legal database, classification system, citation architecture, or structured dataset may be difficult for competitors to reproduce.

The case illustrates that:

ownership of an intellectual-property-related resource does not automatically immunise conduct from competition law.

At the same time, compulsory access is exceptional and requires rigorous analysis.

13. Bronner v Mediaprint

Case C-7/97, Oscar Bronner GmbH & Co. KG v Mediaprint, EU, 1998

Principle

The Court applied a demanding test to refusal-to-supply claims involving an allegedly essential facility.

Relevant considerations included whether:

1. the facility was indispensable;
2. duplication was impossible or economically unreasonable;
3. refusal would eliminate effective competition; and
4. there was no objective justification.

Legal-intelligence relevance

Suppose a dominant legal-data provider controls a dataset that competitors claim they must access.

Bronner cautions against treating every commercially valuable database as an essential facility.

A competitor would generally need to establish genuine indispensability, not merely usefulness.

14. Microsoft Corp. v Commission

Case T-201/04, Microsoft Corp. v Commission, EU, 2007

Principle

The General Court considered Microsoft's refusal to provide interoperability information and its relationship with market foreclosure.

The case is highly significant for understanding:

• interoperability;
• technological ecosystems;
• refusal to supply;
• leveraging dominance;
• network effects.

Legal-intelligence relevance

A legal-intelligence platform could potentially create competitive problems by restricting:

• API access;
• export functionality;
• citation interoperability;
• machine-readable legal data;
• integration with competing AI tools.

The Microsoft case provides an important conceptual framework for examining whether technological restrictions protect legitimate innovation or instead exclude competitors.

15. Google Shopping

Google Search (Shopping), Case AT.39740, European Commission, 2017; General Court judgment in Case T-612/17, 2021

Principle

The European Commission examined Google's preferential positioning of its own comparison-shopping service within general search results.

The case concerned the use of dominance in general search to advantage another service.

Legal-intelligence relevance

A comparable concern could theoretically arise if a dominant legal-search engine systematically preferred its own:

• AI legal assistant;
• case-analysis product;
• legal marketplace;
• legal-research service

over competing services.

The important lesson is that ranking and visibility can become competition issues when controlled by a dominant intermediary.

16. Google Android

Case AT.40099, Google Android, European Commission, 2018; Case T-604/18, Google and Alphabet v Commission, General Court, 2022

Principle

The proceedings examined practices including:

• tying;
• contractual restrictions;
• distribution arrangements;
• ecosystem effects.

Relevance to legal intelligence

A dominant legal-technology ecosystem could potentially use contractual arrangements to extend market power from one product into adjacent legal-information or AI markets.

For example:

Legal database → mandatory AI assistant → mandatory contract-analysis tool

could raise questions about tying or leveraging, depending on the market structure and effects.

17. Aspen Skiing Co. v Aspen Highlands Skiing Corp.

472 U.S. 585, U.S. Supreme Court, 1985

Principle

The U.S. Supreme Court considered circumstances in which a unilateral refusal to deal by a monopolist could violate Section 2 of the Sherman Act.

The case is especially associated with a prior course of cooperation followed by termination that lacked an apparent legitimate business justification.

Legal-intelligence relevance

The case can be relevant where a dominant legal-information platform:

• historically provided data access;
• cooperated with competing services;
• suddenly terminates access;
• and the termination appears to lack a legitimate business justification.

It should not, however, be understood as creating a general duty for dominant firms to deal with competitors.

18. Eastman Kodak Co. v Image Technical Services

504 U.S. 451, U.S. Supreme Court, 1992

Principle

The Supreme Court recognised that substantial market power can sometimes exist in an aftermarket even where competition exists in a primary market.

Relevance to legal intelligence

The principle is useful for analysing situations where a legal-tech company appears competitive at the initial software or subscription stage but later acquires substantial power over:

• proprietary datasets;
• upgrades;
• APIs;
• analytics;
• stored research data;
• specialised modules.

Thus, competition authorities may need to examine aftermarkets and customer lock-in, rather than examining only the initial product market.

19. Additional Important Case: MEO

MEO – Serviços de Comunicações e Multimédia SA v Autoridade da Concorrência, Case C-525/16, EU, 2018

Principle

The Court clarified the assessment of discriminatory pricing under Article 102.

The analysis focuses on whether the conduct is capable of placing trading partners at a competitive disadvantage, taking account of the competitive conditions.

Legal-intelligence relevance

A dominant legal-data platform might charge:

• law firms one price;
• AI developers another;
• competing legal databases substantially higher API fees.

Different prices are not automatically unlawful. The critical question is whether discriminatory conditions distort competition and whether there is objective justification.

20. Competition Issues Specific to Legal AI

21. Market Power Through Legal Citation Networks

Legal citation databases can produce an unusual form of network effect.

A platform that maps:

Case A → Case B → Case C → statutory provision → subsequent treatment

can create a highly valuable legal knowledge graph.

The competitive advantage may come not simply from possessing the underlying judgments, but from:

• classification;
• tagging;
• citation relationships;
• treatment history;
• editorial analysis;
• AI-generated relationships.

Consequently, the competition analysis should distinguish between:

public legal information

and

privately created value-added information infrastructure.

22. Public Data and Competition Law

A particularly important question is whether legal information should be treated differently where the underlying material is public.

Judgments and legislation may be publicly accessible, but digitising and structuring them can require significant investment.

Therefore:

Public availability of the underlying information does not necessarily mean that every privately created database is competitively interchangeable.

However, if competitors can readily obtain the same information from public sources and recreate competing services at reasonable cost, claims of indispensability become substantially weaker.

23. Acquisitions and Legal-Intelligence Market Power

Competition authorities may scrutinise acquisitions where a large legal-information incumbent purchases:

• a promising legal-AI startup;
• a specialised litigation-analytics company;
• a legal citation provider;
• a legal-data aggregator.

Particular attention may be given to:

• nascent competition;
• potential competition;
• access to proprietary datasets;
• AI model capabilities;
• customer foreclosure;
• interoperability;
• innovation competition.

This is especially important where traditional turnover-based merger thresholds may fail to capture the competitive significance of a small but strategically important AI company.

24. Remedies

Potential competition remedies include:

Structural remedies

• divestiture;
• separation of business units.

Behavioural remedies

• non-discriminatory API access;
• data portability;
• interoperability requirements;
• transparent ranking criteria;
• prohibition of tying;
• non-exclusive licensing.

Data-related remedies

• access to specified datasets;
• machine-readable formats;
• portability of customer research histories;
• standardised metadata.

Merger remedies

• licensing commitments;
• firewall arrangements;
• interoperability commitments;
• divestiture of overlapping datasets or products.

25. Indian Competition-Law Application

Under Section 4 of the Competition Act, 2002, the relevant inquiry would potentially involve:

1. defining the relevant product and geographic market;
2. determining whether the legal-intelligence provider is dominant;
3. identifying the conduct;
4. determining whether the conduct falls within an abusive category;
5. assessing actual or potential foreclosure;
6. considering efficiencies and legitimate business justifications.

Potential conduct could include:

• discriminatory access to legal data;
• unfair conditions imposed on subscribers;
• denial of interoperability;
• tying AI products to legal databases;
• exclusionary contractual arrangements;
• leveraging database dominance into AI services;
• predatory pricing;
• discriminatory API access.

26. Hypothetical Example

Assume LegalAI X controls 80% of a market for professional legal research.

It owns a large proprietary database and launches an AI research assistant.

It then:

1. gives its own AI product unrestricted API access;
2. charges competing AI developers extremely high access fees;
3. prevents customers from exporting research histories;
4. ranks its AI answers above competing research services;
5. requires customers purchasing the database to purchase its AI assistant.

The conduct should not automatically be characterised as unlawful merely because the company is dominant.

Each practice would require separate analysis:

The central issue would be whether these practices protect legitimate product quality or instead materially foreclose competing legal-intelligence providers.

27. Key Doctrinal Lessons from the Cases

The cases collectively demonstrate several principles:

1. Dominance is contextual

United Brands demonstrates that economic strength must be evaluated within the competitive structure of the market.

2. Proprietary information can have competition significance

IMS Health illustrates the difficult boundary between intellectual-property rights and competition law.

3. Not every important database is an essential facility

Bronner establishes a demanding threshold for indispensability.

4. Interoperability can become competitively important

Microsoft demonstrates how technological restrictions can interact with dominance.

5. Search rankings can affect competition

Google Shopping illustrates the significance of preferential treatment by a dominant search intermediary.

6. Ecosystem leverage matters

Google Android shows how contractual arrangements can extend market power across connected products.

7. Refusal to deal requires careful analysis

Aspen Skiing demonstrates that the circumstances surrounding termination of an existing commercial relationship can matter.

8. Aftermarkets can generate market power

Kodak shows why competition analysis may need to examine lock-in and secondary markets.

28. Conclusion

Legal intelligence is increasingly capable of becoming a concentrated, data-intensive digital market. Its competitive characteristics differ from traditional legal publishing because market power may arise simultaneously from databases, AI models, citation networks, proprietary metadata, APIs, interoperability, network effects and customer switching costs.

The most important competition-law questions are therefore:

Who controls the legal data? Who can access it? Who controls the interfaces through which it is accessed? Can competitors interoperate? Can customers switch? And does the incumbent use its position in one legal-information market to restrict competition in another?

The jurisprudence of United Brands, IMS Health, Bronner, Microsoft, Google Shopping, Google Android, Aspen Skiing, Kodak, and MEO provides a useful doctrinal foundation for analysing these questions.