Competition Law and Strategic Competition Policy for Intelligent Societies

1. Introduction

An intelligent society is a society in which artificial intelligence, big-data analytics, connected infrastructure, autonomous systems, digital identity, smart cities, algorithmic decision-making and intelligent networks increasingly influence economic activity.

Competition in such societies is therefore not limited to traditional firms selling traditional products. Competitive advantage may depend upon control over:

AI models;

computing infrastructure;

datasets;

digital identity;

cloud services;

algorithms;

smart-city infrastructure;

digital payment systems;

connected devices;

telecommunications;

autonomous systems;

intelligent energy networks; and

interoperability standards.

This creates a major challenge for competition law.

The central question becomes:

How can competition policy preserve innovation, market contestability and consumer choice when intelligent technologies create powerful feedback loops between data, algorithms, infrastructure and market power?

Strategic competition policy must therefore protect dynamic competition, not merely current market shares.

2. Meaning of an Intelligent Society

An intelligent society can be understood as an economic and social environment where decisions and services are increasingly mediated by interconnected intelligent technologies.

Examples include:

Smart cities

intelligent transportation;

automated traffic management;

smart electricity grids;

digital public services.

Intelligent healthcare

AI diagnostics;

health-data platforms;

digital hospitals;

predictive medicine.

Intelligent finance

algorithmic credit assessment;

digital payments;

automated investment;

fraud-detection systems.

Intelligent industry

industrial AI;

robotics;

predictive maintenance;

autonomous manufacturing.

Intelligent commerce

algorithmic pricing;

recommendation systems;

automated advertising;

AI marketplaces.

Consequently, competition may increasingly occur between technology ecosystems, rather than individual products.

3. From Digital Competition to Intelligent-Economy Competition

Traditional competition:

Firm → Product → Consumer

Digital-platform competition:

Platform → Users → Complementors → Consumers

Intelligent-society competition:

Data → AI → Infrastructure → Algorithms → Ecosystem → Users → More Data

This produces a powerful feedback loop:

More users → more data → better algorithms → better services → more users.

The loop can generate legitimate innovation.

However, if an incumbent controls multiple layers, it can also create substantial entry barriers.

4. Strategic Competition Policy

Strategic competition policy involves using competition law to preserve the long-term competitive structure of markets.

It has five major objectives:

contestability;

innovation;

interoperability;

access to critical inputs; and

prevention of durable ecosystem foreclosure.

It combines:

Ex post enforcement

Against:

abuse of dominance;

cartels;

exclusionary agreements;

discriminatory access;

tying;

predatory conduct.

Ex ante governance

Through:

interoperability;

data portability;

merger scrutiny;

access rules;

transparency;

non-discrimination.

Dynamic analysis

Considering:

potential competitors;

innovation pipelines;

technological substitution;

startup entry;

future markets.

5. Strategic Assets in Intelligent Societies

Competition policy must identify assets that can create ecosystem power.

A. Data

Data can improve:

AI training;

prediction;

personalization;

fraud detection;

pricing;

recommendation.

B. Computing Power

AI development can require:

GPUs;

specialized chips;

cloud computing;

data centres.

Control over these inputs may affect downstream competition.

C. Algorithms

Algorithms may determine:

prices;

ranking;

access;

advertising;

credit;

recommendations.

D. Digital Infrastructure

Examples include:

cloud systems;

telecommunications;

payment networks;

digital identity;

smart-grid systems.

E. Standards

Technical standards may determine which technologies are compatible.

6. Network Effects in Intelligent Markets

Intelligent systems can produce unusually strong network effects.

For example:

More consumers

↓

More behavioural data

↓

Better AI

↓

Better recommendations

↓

More consumers

This is a data-network effect.

A competing firm may possess a technically superior algorithm but still struggle because it lacks comparable data.

Competition policy should therefore examine whether:

data is replicable;

consumers can port data;

rivals can access necessary inputs;

multi-homing is possible;

switching costs are significant.

7. Key Case Laws

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

Facts

Microsoft possessed substantial power in operating systems and engaged in conduct affecting browser competition.

Principle

The court examined whether Microsoft's conduct unlawfully maintained its monopoly by restricting avenues through which competitors could develop.

Importance for intelligent societies

The case is foundational for understanding:

network effects;

technological ecosystems;

platform leverage;

innovation;

exclusion of emerging technologies.

It demonstrates that competition law must consider future competitive threats, particularly in technology markets.

8. Terminal Railroad Association of St. Louis, 224 U.S. 383 (1912)

Facts

Railroad companies controlled strategically important infrastructure necessary for competitors to access the St. Louis market.

Principle

Control over a critical bottleneck could restrict competition.

Modern relevance

The underlying principle can inform analysis of:

cloud infrastructure;

telecommunications;

payment systems;

smart-grid infrastructure;

digital identity;

data infrastructure.

The case illustrates that competition problems may arise upstream, before competitors reach consumers.

9. Aspen Skiing Co. v. Aspen Highlands Skiing Corp., 472 U.S. 585 (1985)

Principle

Under particular circumstances, termination of a previous cooperative arrangement by a dominant firm could constitute exclusionary conduct.

Intelligent-society relevance

The principle can be relevant to:

interoperability;

API access;

technical cooperation;

ecosystem participation;

shared infrastructure.

However, the case does not establish that every refusal to cooperate is unlawful.

10. Verizon Communications Inc. v. Trinko, 540 U.S. 398 (2004)

Principle

The Supreme Court was reluctant to impose broad duties upon firms to share their assets with competitors.

Competition-policy significance

Intelligent economies depend heavily upon private investment in:

AI infrastructure;

cloud computing;

data centres;

telecommunications;

advanced semiconductor systems.

Compulsory access must therefore be carefully justified.

Otherwise, regulation may reduce incentives to innovate and invest.

11. Bronner v. Mediaprint, Case C-7/97

Principle

The Court of Justice adopted a demanding standard for requiring a dominant company to provide access to infrastructure.

The concept of indispensability is particularly important.

Intelligent-society relevance

Not every:

AI model;

database;

API;

cloud service;

technological interface

should automatically be treated as an essential facility.

Competition policy must determine whether realistic alternatives exist.

12. IMS Health v. Commission, Case C-418/01 P

Principle

The case concerned access to a protected information structure and the circumstances under which refusal to license intellectual property could raise abuse-of-dominance concerns.

Relevance

The reasoning is important for modern:

proprietary datasets;

software architecture;

APIs;

technological standards;

data interoperability.

It demonstrates the need to balance:

intellectual-property incentives

with

competitive access.

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

Principle

The European Commission's decision, upheld in substantial part, concerned Microsoft's conduct involving interoperability information and tying.

Intelligent-economy relevance

The case is particularly significant because technological interoperability can determine whether competing innovators can participate in an ecosystem.

Potential modern analogues include:

AI APIs;

cloud interoperability;

smart-device ecosystems;

connected-vehicle systems;

digital identity systems.

14. Google Shopping, Commission Decision AT.39740

Competition issue

The European Commission examined Google's treatment of comparison-shopping services within its general search results.

Relevance

The case demonstrates how control over an important digital access point can affect competition in adjacent markets.

In intelligent societies, similar questions can arise where an AI-powered intermediary:

ranks its own services;

controls recommendations;

determines visibility;

uses proprietary data;

directs users toward affiliated services.

15. T-Mobile Netherlands, Case C-8/08

Principle

The case concerned information exchange between competitors in telecommunications.

Information exchange can reduce strategic uncertainty and facilitate coordination.

Intelligent-society significance

AI can make information processing much faster and more sophisticated.

Therefore, competition authorities must examine whether algorithmic systems:

facilitate coordination;

reduce independent decision-making;

transmit competitively sensitive information;

produce coordinated outcomes.

16. Eturas, Case C-74/14

Facts

An online booking platform facilitated a mechanism affecting discount levels offered by participating businesses.

Principle

Digital systems can facilitate coordination between otherwise independent competitors.

Modern significance

The case is particularly relevant to:

algorithmic pricing;

AI marketplaces;

automated bidding;

recommendation systems;

platform-mediated competition.

Technology does not remove competition-law responsibility merely because the coordination occurs through software.

17. Competition Issues in Intelligent Societies

A. AI Ecosystem Dominance

A company may control:

chips → cloud → foundation model → API → applications

This can create vertical ecosystem power.

Competition concerns may arise if the company:

restricts access to computing;

favours its own models;

bundles AI with cloud services;

discriminates against rival developers.

18. Data Concentration

Data concentration can produce significant competitive advantages.

A dominant ecosystem might combine:

search data;

shopping data;

location data;

financial data;

behavioural data;

device data.

This creates the possibility of:

data accumulation → improved AI → stronger market position → further data accumulation.

Competition policy should therefore examine data as a potential strategic competitive input.

19. Intelligent Pricing and Algorithmic Collusion

Algorithms can independently monitor:

competitors' prices;

inventory;

demand;

consumer behaviour.

This can create competition concerns if algorithms facilitate coordinated conduct.

Important distinctions must be maintained between:

Legitimate algorithmic pricing

Each firm independently determines its prices.

Algorithmically facilitated coordination

Competitors use technology in a manner that reduces strategic uncertainty and facilitates coordinated outcomes.

Explicit algorithmic agreement

Competitors directly agree to use algorithms to coordinate conduct.

The legal consequences can differ substantially.

20. AI Self-Preferencing

Suppose an AI assistant controls consumer recommendations.

It might recommend:

its own payment system → its own shopping service → its own financial product

rather than competing services.

The competition question becomes:

Is the recommendation based upon legitimate product quality and efficiency, or is the AI ecosystem using its gatekeeper position to disadvantage competing providers?

This is the intelligent-economy equivalent of traditional self-preferencing concerns.

21. Digital Identity as a Competition Bottleneck

Digital identity systems may become essential to:

banking;

healthcare;

government services;

telecommunications;

e-commerce.

If one provider controls identity verification, it may potentially influence competition downstream.

Relevant questions include:

Can competitors access the identity system?

Are access terms discriminatory?

Is the system interoperable?

Can users switch?

Is the infrastructure duplicable?

22. Smart Cities and Competition Law

Smart-city infrastructure may involve:

transportation;

energy;

telecommunications;

surveillance technology;

digital payments;

public data;

cloud services.

A single infrastructure provider may operate across several layers.

Competition authorities should therefore distinguish:

regulatory monopoly

from

commercial exploitation of monopoly control.

State involvement does not automatically resolve competition concerns.

23. Intelligent Healthcare Markets

AI healthcare ecosystems may combine:

patient data;

diagnostic algorithms;

hospital systems;

insurance;

pharmaceutical data;

cloud computing.

Competition concerns may arise from:

exclusive datasets;

discriminatory access;

tying;

interoperability restrictions;

strategic acquisitions;

data consolidation.

At the same time, healthcare markets require strong privacy and safety protections.

Competition policy must therefore operate consistently with legitimate:

privacy;

cybersecurity;

safety;

medical-regulation objectives.

24. Autonomous Vehicle Ecosystems

An autonomous-vehicle ecosystem may involve:

Vehicle → sensors → mapping → AI → cloud → charging → payment → mobility platform

Competition may become concentrated at several levels.

Potential concerns include:

exclusive mapping data;

charging-network restrictions;

proprietary interfaces;

vehicle-to-infrastructure interoperability;

cloud lock-in;

strategic acquisitions.

Thus, competition analysis should consider the entire technological ecosystem, rather than only vehicle manufacturers.

25. Strategic Acquisitions

Intelligent economies make acquisitions particularly important.

A dominant technology company may acquire:

an AI startup;

a dataset provider;

a semiconductor designer;

a robotics company;

a cybersecurity firm;

an autonomous-driving company.

The acquired firm may have low current revenue but substantial future competitive potential.

Competition authorities should therefore examine:

innovation capability;

pipeline products;

patents;

data;

R&D;

potential entry;

complementary technology.

26. Competition and Innovation Incentives

Strategic competition policy must avoid treating technological success as inherently anticompetitive.

A successful company may legitimately obtain advantages from:

superior technology;

R&D;

economies of scale;

better products;

lower costs;

investment;

innovation.

Competition law becomes concerned where the firm uses that position to artificially prevent competitive alternatives.

The distinction is:

Competition on the merits versus exclusion of competition.

27. India: Competition Act, 2002

The principal provisions include:

Section 3

Addresses anticompetitive agreements.

Relevant to:

algorithmic coordination;

information exchange;

restrictive technology arrangements;

exclusive agreements.

Section 4

Addresses abuse of dominant position.

Potential concerns include:

discriminatory access;

denial of market access;

tying;

leveraging;

unfair conditions.

Sections 5 and 6

Address combinations and merger control.

They become increasingly important for:

AI acquisitions;

data acquisitions;

technology consolidation;

ecosystem mergers.

Section 19

Provides the basis for investigation of competition concerns.

28. Strategic Competition Assessment

A competition authority examining an intelligent ecosystem can use the following framework:

Step 1 — Identify the ecosystem

Who controls:

infrastructure;

data;

AI;

distribution;

standards?

Step 2 — Measure dependence

Who depends on the ecosystem?

consumers;

developers;

suppliers;

startups;

governments.

Step 3 — Examine barriers

Consider:

network effects;

switching costs;

data advantages;

technical incompatibility;

capital requirements.

Step 4 — Examine conduct

Investigate:

tying;

bundling;

self-preferencing;

exclusion;

discriminatory access;

refusal to deal;

algorithmic coordination.

Step 5 — Examine dynamic effects

Ask:

Could the conduct eliminate future competition?

Does it reduce innovation?

Does it prevent entry?

Does it restrict technological alternatives?

Step 6 — Examine legitimate justification

Consider:

security;

privacy;

safety;

technical compatibility;

efficiency;

innovation.

Step 7 — Determine remedy

Possible remedies include:

interoperability;

data portability;

access obligations;

non-discrimination;

behavioural commitments;

merger remedies;

structural remedies in exceptional circumstances.

29. Remedies for Intelligent-Economy Competition Problems

A. Interoperability

Require compatible interfaces between systems.

B. Data Portability

Allow users or businesses to transfer relevant data.

C. Non-Discrimination

Prevent discriminatory treatment of competing providers.

D. Access Remedies

Provide reasonable access to genuinely indispensable infrastructure.

E. Merger Remedies

Require:

divestiture;

licensing;

interoperability;

data-access commitments.

F. Transparency

Require sufficient information concerning:

ranking;

access;

pricing;

automated decisions.

Transparency should not, however, require disclosure of legitimate trade secrets or source code in every circumstance.

30. Case-Law Summary

31. Core Principles of Strategic Competition Policy

The competition-policy framework for intelligent societies can be reduced to ten principles:

Protect competition, not individual competitors.

Assess ecosystems rather than isolated products where appropriate.

Treat data as a potential strategic competitive input.

Examine interoperability and switching costs.

Consider future and potential competition.

Scrutinize strategic technology acquisitions.

Monitor algorithmic coordination.

Distinguish legitimate innovation from exclusionary conduct.

Preserve incentives for investment and technological development.

Use proportionate remedies directed at the specific competitive harm.

32. Conclusion

Competition law in intelligent societies must evolve from a primarily firm-centric model toward a dynamic ecosystem-oriented model.

The critical sources of market power may increasingly be:

AI;

data;

algorithms;

computing capacity;

infrastructure;

digital identity;

interoperability;

standards;

network effects; and

ecosystem integration.

The leading cases—Terminal Railroad, Aspen Skiing, Trinko, Bronner, IMS Health, Microsoft, Google Shopping, T-Mobile Netherlands and Eturas—provide the doctrinal foundations for addressing these challenges.

The central policy balance is particularly important. Competition law should not penalize firms merely because they develop superior technologies or successful intelligent ecosystems. At the same time, ecosystem leaders should not be permitted to transform technological success into permanent control over future innovation through artificial foreclosure, discriminatory access, interoperability restrictions, strategic acquisitions or algorithmic coordination.

Ultimately, strategic competition policy for intelligent societies seeks to ensure that AI-driven growth, data-driven innovation and intelligent infrastructure remain open enough for new technologies, startups and competing ecosystems to emerge and challenge established market power.