COMPETITION LAW AND LONG-TERM COMPETITION GOVERNANCE FOR INTELLIGENT MARKETS

1. Introduction

Intelligent markets are markets increasingly influenced by artificial intelligence (AI), machine learning, algorithms, automated decision-making, big data, cloud computing, digital platforms, robotics and other advanced technologies.

Examples include:

AI markets;

automated financial markets;

algorithmic advertising;

digital marketplaces;

autonomous transportation;

smart manufacturing;

AI-powered healthcare;

cloud-computing markets;

intelligent logistics;

fintech platforms; and

machine-learning services.

These markets can produce substantial benefits through:

lower transaction costs;

faster decision-making;

better forecasting;

personalised services;

automated production;

improved resource allocation; and

technological innovation.

However, intelligent markets also create long-term competition concerns. Algorithms can reinforce market power, data can become a strategic asset, AI infrastructure can be concentrated, and digital ecosystems can make switching difficult.

Long-term competition governance therefore refers to the continuing legal, regulatory and institutional framework used to preserve competitive markets as technology develops.

2. Meaning of Long-Term Competition Governance

Long-term competition governance is broader than investigating individual violations.

It involves:

identifying emerging competition risks;

monitoring market concentration;

supervising dominant platforms;

examining acquisitions;

protecting innovation;

regulating potentially exclusionary conduct;

monitoring algorithms;

encouraging interoperability;

protecting access to important infrastructure;

maintaining effective market entry; and

adapting competition policy as technology changes.

The objective is to maintain a competitive environment over time, rather than merely correcting an infringement after market power has already become entrenched.

3. Meaning of Intelligent Markets

An intelligent market is a market in which technological systems significantly influence:

pricing;

production;

allocation;

matching;

advertising;

investment;

distribution;

consumer recommendations; or

strategic decision-making.

For example:

An AI-based marketplace automatically determines prices, recommends products, ranks sellers and allocates customers.

The technology may increase efficiency, but it may also influence competitive conditions.

4. Why Intelligent Markets Create New Competition Issues

A. Algorithmic Decision-Making

Algorithms can automatically determine:

prices;

discounts;

rankings;

advertising;

supply allocation;

credit decisions; and

product recommendations.

This creates a challenge for competition authorities because the decision-making process may be complex or difficult to understand.

B. Data Concentration

AI systems often require large quantities of data.

A company with access to:

large datasets;

high-quality data;

real-time data; and

historical data

may develop better algorithms than smaller competitors.

This can create a self-reinforcing advantage:

More users → more data → better AI → better service → more users.

5. Network Effects

Intelligent markets often have strong network effects.

For example, a digital platform may become more valuable as:

more consumers join;

more sellers participate;

more data becomes available; and

more complementary services are developed.

Network effects may create a feedback loop that makes entry increasingly difficult.

6. Economies of Scale

AI and digital technology may involve:

high initial R&D costs;

expensive computing infrastructure;

specialised chips;

large datasets; and

highly skilled employees.

Once the infrastructure exists, however, the technology may be deployed to millions of users at relatively low marginal cost.

This can favour large firms and increase market concentration.

7. Computing Infrastructure as a Competition Issue

Advanced AI systems may depend upon:

cloud computing;

GPUs;

specialised AI chips;

data centres;

model-training infrastructure; and

high-speed networks.

If access to these resources becomes concentrated, emerging competitors may face significant barriers to entry.

Long-term competition governance may therefore need to examine competition at several levels:

chips → computing → foundation models → applications → distribution.

8. Foundation Models and AI Ecosystems

Foundation models can support multiple applications.

A company controlling a major model may expand into:

search;

office software;

advertising;

cloud computing;

coding;

autonomous systems; and

consumer applications.

This creates potential ecosystem concerns.

The company may have incentives to:

favour its own applications;

bundle services;

restrict access;

impose exclusive arrangements;

use data advantages; or

disadvantage competing developers.

9. Market Definition in Intelligent Markets

Market definition can be difficult because intelligent technologies evolve rapidly.

Competition authorities may consider:

functionality;

quality;

price;

data;

privacy;

innovation;

interoperability;

technological substitutability;

geographic scope; and

potential competition.

A market that appears narrow today may become much broader as technology converges.

10. Potential Competition

Potential competition is particularly important in intelligent markets.

A small AI company may currently have:

limited customers;

limited revenue;

limited market share.

However, it may possess:

important patents;

innovative technology;

talented researchers;

proprietary models;

valuable datasets; or

a promising product pipeline.

Its future competitive significance may therefore exceed its current market position.

11. Killer Acquisitions

A dominant technology company may acquire an emerging firm before it becomes a significant competitor.

Competition authorities may examine whether the acquisition:

eliminates a future competitor;

reduces innovation;

removes an alternative technology;

increases data concentration; or

strengthens ecosystem power.

Long-term governance therefore requires merger analysis to consider more than current revenue.

12. Self-Preferencing

Self-preferencing occurs where a platform favours its own services over competing services.

For example:

An AI marketplace gives its own AI assistant higher visibility than independent AI applications.

Possible effects include:

reduced visibility for competitors;

reduced customer choice;

increased dependency on the platform; and

reinforcement of market power.

13. Tying and Bundling

An intelligent-platform operator may require customers to purchase one service together with another.

Examples:

AI software + cloud services;

operating system + AI assistant;

search + advertising;

AI model + application programming interface.

Tying may create efficiencies, but it may also enable a dominant firm to extend market power into another market.

14. Exclusive Agreements

Long-term exclusive agreements may concern:

data;

cloud infrastructure;

computing capacity;

AI distribution;

advertising;

software;

content;

hardware; or

customers.

Exclusivity may encourage investment but can become problematic if it prevents rival firms from obtaining sufficient access to important inputs or customers.

15. Algorithmic Pricing

Algorithms can automatically change prices according to:

demand;

supply;

competitor prices;

customer behaviour;

inventory;

location; and

market conditions.

Algorithmic pricing itself is not unlawful.

The competition concern arises where algorithms:

implement an agreement;

facilitate coordination;

exchange sensitive information;

monitor competitors for coordination; or

are deliberately designed to produce anti-competitive outcomes.

16. Algorithmic Collusion

Intelligent markets may make coordination easier because algorithms can:

monitor prices continuously;

detect competitor changes;

respond immediately; and

adjust prices automatically.

However, parallel pricing does not automatically establish unlawful collusion.

Authorities normally need to distinguish:

independent algorithmic behaviour

from

algorithmically facilitated coordination or agreement.

17. Data as a Source of Market Power

Data may provide competitive advantages through:

improved prediction;

better recommendations;

targeted advertising;

fraud detection;

personalised services;

AI training; and

demand forecasting.

Competition concerns may arise where a dominant firm controls data that rivals cannot realistically obtain or reproduce.

18. Data Access and Interoperability

Long-term competition governance may examine whether users and businesses can:

transfer data;

connect to alternative services;

use competing applications;

access APIs;

interoperate with other platforms.

Technical restrictions may create significant switching costs.

19. Switching Costs

Intelligent markets may create switching costs through:

proprietary data;

customised AI systems;

technical integration;

long-term contracts;

specialised software;

training investment;

customer histories; and

ecosystem dependence.

High switching costs may discourage users from moving to competing services.

20. Digital Ecosystems

An intelligent-market ecosystem may include:

Hardware → Operating System → Cloud → AI Model → Application → Marketplace → Advertising

Control over several levels can create competitive advantages.

Competition authorities may therefore examine whether a company uses power at one level to restrict competition at another.

21. Vertical Foreclosure

Vertical foreclosure occurs where a company uses control over an upstream or downstream market to disadvantage competitors.

Examples:

restricting access to computing capacity;

refusing interoperability;

tying cloud services to AI applications;

restricting access to distribution;

imposing exclusive contracts.

The assessment depends on market power and competitive effects.

22. Intellectual Property

AI and intelligent markets frequently depend on:

patents;

copyrights;

trade secrets;

algorithms;

datasets; and

proprietary technology.

Intellectual-property rights can encourage innovation.

However, competition concerns may arise if IP is used to:

exclude competitors;

prevent interoperability;

impose discriminatory licensing;

restrict technological development; or

prevent legitimate follow-on innovation.

23. Standard-Setting

Technical standards can promote:

interoperability;

safety;

compatibility;

efficiency; and

innovation.

However, standard-setting can also create risks if participants use the process to:

exclude competitors;

discriminate against technologies;

manipulate patent rights; or

prevent alternative standards.

24. Long-Term Merger Governance

Competition authorities should monitor transactions involving:

AI companies;

cloud providers;

semiconductor companies;

data businesses;

digital platforms;

robotics companies; and

other emerging technologies.

Relevant questions include:

Does the target represent potential competition?

Does it possess important technology?

Does it control valuable data?

Will the transaction reduce innovation?

Will the transaction strengthen network effects?

Will competitors lose access to infrastructure?

Will the transaction increase ecosystem power?

25. Important Case Laws

Because intelligent-market competition is a relatively developing field, many important principles come from digital-platform, technology, innovation, intellectual-property and algorithmic-market cases that provide useful analogies.

Case 1: United States v. Microsoft Corp., 253 F.3d 34 (D.C. Cir. 2001)

Facts

Microsoft was found to have engaged in exclusionary conduct involving its Windows operating system and Internet Explorer.

Principle

A dominant technology firm cannot use exclusionary conduct to protect its position by restricting competing technologies.

Relevance to intelligent markets

The case demonstrates the importance of preserving technological pathways for future competitors.

26. Case 2: Ohio v. American Express Co., 585 U.S. 529 (2018)

Principle

Two-sided platforms require analysis that recognises interactions between different sides of the platform.

Intelligent-market relevance

AI marketplaces, advertising platforms and digital ecosystems may connect:

consumers;

developers;

advertisers;

suppliers; and

service providers.

Competition effects may therefore need to be examined across the interconnected platform.

27. Case 3: American Needle, Inc. v. National Football League, 560 U.S. 183 (2010)

Principle

Participants in a common organisation may remain separate economic actors for competition-law purposes.

Intelligent-market relevance

Technology companies may cooperate through:

standards;

platforms;

APIs;

joint ventures; or

technical infrastructure

while continuing to compete with each other.

28. Case 4: Aspen Skiing Co. v. Aspen Highlands Skiing Corp., 472 U.S. 585 (1985)

Principle

In exceptional circumstances, a dominant firm's refusal to cooperate can constitute exclusionary conduct.

Intelligent-market relevance

The case provides an important reference point for disputes involving:

platform access;

interoperability;

essential inputs; and

refusal to deal.

29. Case 5: Verizon Communications Inc. v. Trinko, 540 U.S. 398 (2004)

Principle

Competition law does not generally impose a broad duty on a company to assist competitors.

Intelligent-market relevance

A dominant AI or digital company is not automatically required to provide competitors with its technology or infrastructure. Any access obligation must arise from the applicable competition-law framework and circumstances.

30. Case 6: Google Shopping, European Commission, Case AT.39740

Principle

A dominant digital platform can face competition scrutiny where it favours its own service through control over an important digital gateway.

Intelligent-market relevance

Self-preferencing could become significant where AI platforms control:

search;

recommendations;

application distribution; or

digital marketplaces.

31. Case 7: Google Android, European Commission, Case AT.40099

Principle

Bundling and contractual restrictions within a dominant digital ecosystem can raise competition concerns where they strengthen market power and restrict competing services.

Intelligent-market relevance

The case provides useful principles for analysing AI ecosystems that combine:

operating systems;

cloud services;

AI assistants;

applications; and

distribution.

32. Case 8: FTC v. Actavis, Inc., 570 U.S. 136 (2013)

Principle

Competition analysis may consider the economic substance and potential competitive effects of an agreement rather than merely its formal legal classification.

Intelligent-market relevance

This principle can inform assessment of:

technology licensing;

patent agreements;

AI partnerships;

restrictions on innovation; and

agreements affecting future market entry.

33. Case 9: Intel Corp. v. European Commission, Case C-413/14 P

Principle

The assessment of exclusionary rebates by a dominant company may require consideration of their competitive effects.

Intelligent-market relevance

AI and digital platforms may use:

rebates;

discounts;

cloud credits;

loyalty incentives; or

preferential commercial terms.

Such practices may require careful competition analysis where a dominant firm uses them to exclude efficient rivals.

34. Case 10: FTC v. Illumina, Inc.

The transaction involving Illumina and GRAIL demonstrated the importance of analysing competition involving emerging technologies and potential future competition.

Principle

A transaction can raise competition concerns even when the target is an emerging technology company rather than a traditional large incumbent.

Intelligent-market relevance

AI, robotics, biotechnology and other emerging technologies can similarly involve small firms with significant future competitive potential.

35. Indian Competition-Law Perspective

In India, intelligent-market competition issues primarily arise under the Competition Act, 2002.

Important provisions include:

Section 3

Deals with anti-competitive agreements.

Section 4

Deals with abuse of dominant position.

Sections 5 and 6

Deal with combinations and merger control.

Section 19

Provides the framework for inquiries by the Competition Commission of India.

Digital and intelligent businesses may therefore face scrutiny involving:

self-preferencing;

exclusivity;

tying;

bundling;

discriminatory access;

data advantages;

algorithmic pricing;

acquisitions; and

platform restrictions.

36. Indian Digital-Platform Case: Samir Agarwal v. ANI Technologies

This case concerned competition allegations involving app-based cab platforms.

Importance

The matter illustrates the application of competition concepts to:

algorithmic pricing;

digital platforms;

network effects;

platform participation; and

relationships between platform operators and service providers.

Long-term significance

It demonstrates why competition analysis must adapt to technology-driven markets.

37. Long-Term Governance Framework

A comprehensive competition-governance system for intelligent markets should include several layers.

Layer 1 — Market Monitoring

Authorities should monitor:

concentration;

entry;

innovation;

acquisitions;

data accumulation; and

platform expansion.

Layer 2 — Merger Review

Transactions should be assessed for:

potential competition;

innovation effects;

data concentration;

network effects; and

ecosystem effects.

Layer 3 — Conduct Regulation

Authorities should monitor:

exclusion;

self-preferencing;

tying;

exclusivity;

discriminatory access;

algorithmic coordination.

Layer 4 — Infrastructure Access

Where legally appropriate, competition policy may address access to:

cloud infrastructure;

computing resources;

interoperability;

technical standards; and

other strategically important infrastructure.

Layer 5 — Innovation Protection

Authorities should consider whether market practices:

discourage R&D;

eliminate alternative technologies;

restrict start-ups; or

reduce future technological competition.

38. Role of Algorithmic Auditing

Competition governance may increasingly require examination of algorithms.

Possible areas include:

pricing algorithms;

recommendation systems;

ranking algorithms;

advertising systems;

matching systems;

automated rebates.

An audit may examine whether an algorithm:

discriminates against competitors;

facilitates coordination;

creates unjustified exclusion;

favours affiliated services; or

reinforces an existing market position.

39. Human Oversight

Intelligent markets should not necessarily be governed exclusively through automated systems.

Human oversight can be important for:

competition compliance;

algorithm design;

pricing;

strategic acquisitions;

platform access;

data use; and

risk assessment.

This is particularly important where automated decisions can have significant competitive consequences.

40. Competition Governance and Consumer Welfare

Consumer welfare in intelligent markets may involve:

price;

quality;

privacy;

security;

innovation;

choice;

transparency;

interoperability; and

service reliability.

A free AI or digital service may still raise competition concerns if consumers have progressively fewer meaningful alternatives.

41. Short-Term Versus Long-Term Governance

42. Practical Example

Suppose Company A operates a major AI ecosystem.

It controls:

cloud computing;

a foundation AI model;

an AI application store;

an advertising platform; and

a large consumer database.

Company A begins:

giving its own AI applications preferential ranking;

requiring developers to use its payment system;

offering discounts only to customers using its cloud;

acquiring promising AI start-ups;

restricting interoperability with competing AI systems.

Each practice must be analysed separately.

However, long-term competition governance asks whether the combined effect is to create an ecosystem from which competitors cannot realistically emerge.

Possible concerns include:

self-preferencing;

tying;

exclusivity;

foreclosure;

data concentration;

elimination of potential competitors;

interoperability restrictions; and

reinforcement of network effects.

43. Challenges for Competition Authorities

A. Technological uncertainty

Authorities cannot always predict which technology will succeed.

B. Rapid market development

Competitive conditions can change quickly.

C. Measuring innovation

Innovation is difficult to quantify.

D. Algorithmic complexity

Decision-making systems may be technically difficult to investigate.

E. Data valuation

The competitive value of data may be uncertain.

F. Global markets

Large technology firms operate across multiple jurisdictions.

G. Ecosystem complexity

A single practice may affect several interconnected markets.

44. Difference Between Innovation and Anti-Competitive Conduct

A successful innovation or large market share is not itself evidence of an antitrust violation.

45. Compliance Measures for Intelligent-Market Businesses

Companies should establish a long-term competition programme covering:

merger review;

algorithmic competition review;

data governance;

platform-access policies;

exclusivity review;

rebate and discount policies;

interoperability;

IP licensing;

standard-setting;

competitor information;

acquisition of start-ups; and

periodic reassessment of market power.

46. Quick Revision Points

Remember:

Intelligent Markets = AI + Data + Algorithms + Networks + Platforms + Automation

Main long-term competition concerns:

Market concentration

Network effects

Data concentration

Algorithmic collusion

Self-preferencing

Tying

Bundling

Exclusivity

Interoperability

Platform access

Killer acquisitions

Innovation suppression

Ecosystem leverage

Cloud and computing concentration

Potential competition

47. Key Case-Law Revision Table

48. Exam-Ready Answer

Competition law and long-term competition governance for intelligent markets concerns the development of competition principles capable of preserving competitive conditions in markets influenced by AI, algorithms, data, digital platforms, automation and advanced computing.

Intelligent markets create distinctive concerns because of network effects, economies of scale, data concentration, algorithmic decision-making, switching costs, ecosystem power and rapid technological change.

Major issues include abuse of dominance, self-preferencing, tying, bundling, exclusive agreements, algorithmic coordination, data concentration, interoperability restrictions, vertical foreclosure, potential competition, killer acquisitions and innovation suppression.

Important cases such as United States v. Microsoft, Ohio v. American Express, American Needle, Aspen Skiing, Verizon v. Trinko, Google Shopping, Google Android, FTC v. Actavis, Intel and FTC v. Illumina provide useful principles for analysing technology platforms, market access, exclusionary conduct, two-sided markets, innovation and potential competition.

In India, the Competition Act, 2002 provides the principal legal framework for anti-competitive agreements, abuse of dominance and combinations, while digital-platform cases demonstrate the adaptation of competition law to technology-based markets.

Conclusion

Long-term competition governance for intelligent markets requires a continuing approach rather than a purely reactive model. Competition authorities must consider not only present prices and market shares but also future innovation, potential competitors, data accumulation, network effects, ecosystem expansion and technological access.

The central objective is to allow intelligent technologies to develop while maintaining meaningful opportunities for entry, innovation, interoperability and consumer choice.

For examination purposes, remember:

AI + Algorithms + Data + Network Effects + Ecosystems + Potential Competition + Innovation = Core Issues of Long-Term Competition Governance in Intelligent Markets.