Competition Law and Strategic Foresight for Advanced Intelligence Markets

1. Introduction

Strategic foresight for advanced intelligence markets refers to the use of competition-law analysis to anticipate how emerging technologies—particularly artificial intelligence (AI), foundation models, autonomous systems, advanced analytics, quantum technologies, synthetic data, and intelligent digital platforms—may alter market structure before conventional antitrust problems become entrenched.

Competition law traditionally examines existing markets, conduct, and transactions. Advanced-intelligence markets require an additional forward-looking dimension because competitive advantages may arise from control over:

• computational infrastructure and specialised chips;
• foundation models and training capabilities;
• high-quality datasets;
• cloud infrastructure;
• AI talent and technical know-how;
• application-programming interfaces (APIs);
• distribution channels and operating systems;
• user and behavioural data;
• proprietary standards;
• model marketplaces and ecosystems;
• interoperability and switching infrastructure.

Strategic foresight therefore does not mean predicting which company will succeed. It means identifying foreseeable mechanisms through which technological development could create, strengthen, or preserve market power.

2. Meaning of Strategic Foresight in Competition Law

Strategic foresight is a structured approach to asking:

How might today's technological or commercial decisions change the competitive structure of tomorrow's market?

In advanced-intelligence markets, regulators may need to examine not merely present market shares but also:

1. future entry conditions;
2. control of essential inputs;
3. network effects;
4. economies of scale and scope;
5. data advantages;
6. vertical integration;
7. interoperability barriers;
8. ecosystem dependencies;
9. acquisitions of potential competitors; and
10. the ability of incumbents to influence technological standards or future market architecture.

This is particularly important where a market is still developing and current market shares may not accurately reveal future competitive constraints.

3. Why Advanced Intelligence Markets Create Special Antitrust Problems

A. Rapid technological development

AI and related technologies can evolve faster than conventional market-definition exercises.

A firm with a relatively small position today may control an important technological bottleneck tomorrow.

B. Economies of scale

Training sophisticated models can require enormous quantities of:

• computing power;
• data;
• engineering resources;
• capital; and
• specialised expertise.

These economies can produce substantial barriers to entry.

C. Data advantages

Large-scale access to data can improve:

• model training;
• personalisation;
• prediction;
• recommendation;
• fraud detection;
• search;
• advertising;
• autonomous decision-making.

The competitive concern arises where accumulated data advantages become difficult for rivals to replicate.

D. Network effects

The value of an intelligent platform may increase as more users, developers, advertisers, or enterprises participate.

This can create a feedback loop:

Users → Data → Better AI → More Users → More Data

Such feedback mechanisms may strengthen incumbent positions.

4. Strategic Foresight and Market Definition

Traditional market definition asks what products or services constrain one another.

In advanced-intelligence markets, regulators may need to examine several layers simultaneously.

Layer 1 — Infrastructure

Examples:

• GPUs;
• AI accelerators;
• cloud computing;
• data centres;
• specialised networking.

Layer 2 — Intelligence

Examples:

• foundation models;
• large language models;
• multimodal models;
• AI agents;
• specialised models.

Layer 3 — Distribution

Examples:

• operating systems;
• search engines;
• app stores;
• enterprise software;
• cloud marketplaces.

Layer 4 — Applications

Examples:

• AI medical diagnostics;
• financial AI;
• autonomous vehicles;
• legal AI;
• industrial AI;
• education platforms.

Layer 5 — Data and feedback

Data generated at each level may reinforce competitive advantages at other levels.

Thus, an apparently competitive application market could depend upon a highly concentrated upstream intelligence or infrastructure market.

5. Strategic Foresight and Market Power

Market power in advanced-intelligence markets may be derived from several complementary assets.

The key question is therefore not simply:

"Who has the largest market share?"

It is:

"Which resources or control points could allow a firm to influence the competitive trajectory of the market?"

6. Strategic Foresight and Merger Control

Mergers are particularly important because an acquisition can eliminate a competitive threat before the target becomes a major competitor.

This creates several possible concerns.

A. Killer acquisitions

An incumbent may acquire an emerging firm possessing:

• innovative AI technology;
• specialised datasets;
• talented researchers;
• a promising model;
• complementary infrastructure.

The target's current revenues may be small even though its future competitive significance could be substantial.

B. Nascent competition

A startup may currently constrain an incumbent through innovation rather than market share.

Strategic merger analysis therefore considers:

• R&D pipelines;
• technological capabilities;
• customer switching;
• potential future products;
• internal documents;
• hiring patterns;
• development roadmaps.

C. Vertical acquisitions

A cloud provider acquiring an AI model developer could potentially combine:

Compute + Model + Distribution

Similarly:

Operating System + AI Assistant + Search + Advertising

may create opportunities for leveraging across markets.

7. Strategic Foresight and Data Concentration

Data concentration can have several competitive effects.

Positive effects

Data aggregation may create:

• improved model performance;
• better products;
• lower costs;
• innovation;
• personalised services.

Potential competition concerns

The same concentration may:

• increase entry barriers;
• make rivals dependent upon incumbent datasets;
• reduce switching;
• facilitate exclusion;
• reinforce network effects.

A competition authority therefore needs to distinguish between legitimate data economies and strategic conduct that prevents rivals from obtaining competitively necessary inputs.

8. Strategic Foresight and AI Ecosystems

Advanced-intelligence markets frequently operate as ecosystems.

For example:

Cloud → Compute → Foundation Model → API → Applications → Users → Data

A firm controlling several layers may possess opportunities to favour its own downstream services.

Potential practices include:

• preferential API access;
• discriminatory pricing;
• tying;
• bundling;
• self-preferencing;
• exclusive agreements;
• interoperability restrictions;
• technical degradation of competing products.

The competitive assessment should examine whether the conduct merely reflects technical integration or instead restricts effective competition.

9. Strategic Foresight and Essential Facilities

Some advanced-intelligence inputs may acquire bottleneck characteristics.

Potential examples include:

• specialised computing infrastructure;
• highly specialised datasets;
• interoperability interfaces;
• technical standards;
• critical AI marketplaces.

The essential-facilities doctrine is applied cautiously because requiring access can reduce incentives to invest.

A foresight-based analysis should therefore ask:

1. Is the input genuinely difficult to reproduce?
2. Is access technically feasible?
3. Is the input indispensable for effective competition?
4. Does denial exclude competitors?
5. Is there an objective justification?
6. Can access remedies preserve investment incentives?

10. Strategic Foresight and Algorithmic Coordination

AI systems can potentially change the nature of coordination between competitors.

Competition authorities may examine whether algorithms:

• independently optimise prices;
• learn from competitors' behaviour;
• facilitate tacit coordination;
• implement common pricing strategies;
• increase transparency beyond competitive levels.

The important distinction is between:

legitimate independent algorithmic optimisation

and

algorithmically facilitated coordination resulting from an anticompetitive agreement or concerted practice.

The mere use of AI does not itself establish an infringement.

11. Strategic Foresight and Autonomous Decision-Making

Advanced intelligence can increasingly make commercial decisions without direct human intervention.

Examples include:

• dynamic pricing;
• inventory management;
• advertising allocation;
• credit decisions;
• procurement;
• logistics;
• recommendation systems.

Competition law may therefore need to examine:

Who designed the algorithm, who controlled it, what information it received, and whether competitors coordinated through it?

Responsibility can become more complicated where pricing or allocation decisions are produced through autonomous systems.

12. Strategic Foresight and Interoperability

Interoperability can be critical in emerging intelligence ecosystems.

A dominant firm could potentially restrict:

• API access;
• data portability;
• model portability;
• compatibility;
• plug-ins;
• third-party applications.

Such restrictions may increase switching costs.

Competition analysis should distinguish:

genuine security or technical requirements

from

artificial restrictions designed to preserve ecosystem power.

13. Strategic Foresight and Consumer Lock-In

Advanced intelligence products can become deeply integrated into business operations.

For example:

AI assistant → enterprise data → workflow → cloud → applications → employee productivity

Once integrated, switching may become expensive.

Potential sources of lock-in include:

• proprietary data formats;
• non-portable models;
• contractual restrictions;
• accumulated user preferences;
• proprietary APIs;
• workflow dependencies.

Strategic foresight asks whether today's integration will become tomorrow's barrier to entry.

14. Strategic Foresight and Innovation Competition

Innovation competition is particularly important in advanced-intelligence markets.

A firm may compete not primarily through current prices but through:

• better models;
• lower inference costs;
• improved accuracy;
• new architectures;
• greater safety;
• specialised applications.

Consequently, competition authorities may need to examine innovation pipelines and not merely current sales.

15. Important Case Laws

Case 1 — United States v. Microsoft Corp. (2001)

The Microsoft litigation is a foundational authority concerning technological ecosystems and leveraging of market power.

Microsoft possessed substantial power in the PC operating-system market and was found liable for various exclusionary practices concerning the browser market.

Relevance to advanced intelligence

The case illustrates how control of one technological platform can be leveraged into adjacent markets.

The foresight lesson is:

Control over a foundational technological layer can become a mechanism for controlling emerging complementary markets.

For AI, comparable analytical questions can arise where a firm controls an operating system, cloud platform, search service, or other distribution channel through which competing AI products reach users.

Case 2 — United States v. Google LLC — Search (D.D.C. 2024)

The Google search litigation concerns alleged exclusionary agreements affecting distribution of general search services.

Relevance

The case demonstrates the importance of distribution bottlenecks.

Even where competing technology can technically be developed, competition may be weakened if an incumbent controls important channels through which users encounter competing products.

Strategic foresight lesson

AI competition may depend not merely upon who develops the strongest model, but upon who controls:

• search;
• browsers;
• operating systems;
• mobile distribution;
• enterprise software; and
• default settings.

Case 3 — United States v. Google LLC — Ad Tech

The U.S. government's advertising-technology litigation illustrates concerns associated with vertical integration across multiple layers of a digital ecosystem.

Relevance

An integrated firm may operate across:

• technology infrastructure;
• exchanges;
• publisher services;
• advertiser services.

The competitive concern is whether vertical integration enables exclusion or discriminatory treatment of rivals.

Strategic foresight lesson

Similar reasoning can become relevant to AI ecosystems where one undertaking controls:

infrastructure + model + marketplace + downstream application.

Case 4 — FTC v. Meta Platforms, Inc.

The FTC's litigation concerning Meta's acquisitions of Instagram and WhatsApp focuses on competition and acquisitions of important emerging platforms.

Relevance

The case illustrates the difficulty of assessing acquisitions involving products whose future competitive significance may exceed their present market position.

Strategic foresight lesson

Merger control in advanced-intelligence markets should consider:

• potential competition;
• innovation pipelines;
• emerging technologies;
• future ecosystem competition; and
• whether an acquisition removes an important future competitive constraint.

Case 5 — European Commission v. Google (Google Shopping)

The European Commission's Google Shopping decision concerned preferential treatment of Google's comparison-shopping service within general search results.

Relevance

The case is important for understanding self-preferencing by a platform that operates both as an infrastructure provider and as a downstream competitor.

Strategic foresight lesson

In AI ecosystems, similar questions may arise where an infrastructure or platform operator:

• hosts competing applications;
• controls ranking or visibility;
• provides AI services itself; and
• has access to information concerning rival users.

Case 6 — Google Android — European Commission

The European Commission's Android decision concerned contractual practices involving Google's mobile operating-system ecosystem.

Relevance

The case illustrates how contractual arrangements and distribution restrictions can reinforce platform power.

Strategic foresight lesson

AI providers may increasingly depend upon:

• mobile operating systems;
• app stores;
• cloud platforms;
• enterprise software ecosystems.

Contractual restrictions affecting access to these ecosystems may therefore have consequences beyond the immediate contractual relationship.

Case 7 — European Commission v. Microsoft (Microsoft II)

The European Commission's Microsoft proceedings included concerns regarding interoperability and tying involving Microsoft's software ecosystem.

Relevance

Interoperability can determine whether competing products can effectively operate alongside a dominant platform.

Strategic foresight lesson

In advanced intelligence markets, interoperability may concern:

• model APIs;
• AI agents;
• enterprise systems;
• cloud environments;
• data formats;
• software tools.

A firm should not necessarily be required to make every proprietary technology open, but interoperability restrictions can become relevant where they materially foreclose competition.

Case 8 — United States v. Apple Inc. (2024)

The U.S. Department of Justice's antitrust action against Apple addresses alleged exclusionary conduct involving Apple's ecosystem.

Relevance

The case highlights the importance of examining an ecosystem rather than evaluating isolated products independently.

Strategic foresight lesson

Competition authorities examining advanced intelligence may need to study how:

hardware + operating system + applications + data + AI services

interact to produce durable competitive advantages.

16. Comparative Case-Law Principles

17. Strategic Foresight Framework

A competition authority can use a six-stage framework.

Stage 1 — Identify technological dependencies

Map:

Compute → Data → Models → APIs → Distribution → Applications

Stage 2 — Identify control points

Determine which firms control:

• scarce inputs;
• technical standards;
• infrastructure;
• distribution;
• data;
• interfaces.

Stage 3 — Identify feedback loops

For example:

More users → More data → Better model → Better service → More users

Stage 4 — Identify exclusion mechanisms

Potential mechanisms include:

• tying;
• bundling;
• exclusivity;
• self-preferencing;
• discriminatory access;
• interoperability restrictions;
• predatory conduct;
• acquisitions.

Stage 5 — Test alternative futures

Regulators can examine scenarios such as:

Scenario A: Several competing foundation-model providers.

Scenario B: Two or three vertically integrated AI ecosystems.

Scenario C: One infrastructure layer becomes highly concentrated.

Scenario D: Open-source models substantially reduce entry barriers.

Stage 6 — Select proportionate intervention

Possible responses include:

• behavioural remedies;
• interoperability requirements;
• access obligations;
• merger remedies;
• monitoring;
• data portability;
• structural remedies in exceptional circumstances.

18. Strategic Foresight and Remedies

Remedies should address the identified competitive mechanism rather than technological development itself.

A. Interoperability remedies

Require technical compatibility where justified.

B. Data portability

Allow users or businesses to transfer relevant data.

C. Non-discrimination

Prevent discriminatory access to infrastructure or interfaces.

D. Structural separation

In exceptional circumstances, separate incompatible commercial functions.

E. Merger remedies

Possible remedies include:

• divestitures;
• licensing;
• access commitments;
• restrictions on information exchange.

F. Monitoring

Technology markets may require continuous monitoring because competitive conditions can change rapidly.

19. Risks of Excessive Strategic Foresight

Strategic foresight must also have limits.

1. False positives

Authorities may incorrectly assume that a technology will become dominant.

2. Innovation chilling

Over-regulation can discourage investment.

3. Dynamic efficiency

A temporary advantage may reflect legitimate innovation rather than exclusion.

4. Difficult prediction

Technological trajectories are uncertain.

5. Remedy risks

An inappropriate access obligation can reduce incentives to develop new technologies.

Therefore, foresight should be based on evidence, technological facts, economic analysis, and demonstrable competitive mechanisms, rather than speculation.

20. Key Competition-Law Questions for Advanced Intelligence Markets

A regulator should ask:

1. Who controls the critical computational resources?
2. Can competitors obtain comparable computing capacity?
3. Is the relevant data replicable?
4. Are users able to switch AI providers?
5. Are APIs interoperable?
6. Does the platform favour its own AI services?
7. Are exclusivity agreements preventing entry?
8. Could an acquisition eliminate a future competitor?
9. Are algorithms facilitating coordination?
10. Does vertical integration create foreclosure opportunities?
11. Are network effects creating durable entry barriers?
12. Does the conduct reduce innovation competition?
13. Are claimed efficiencies verifiable?
14. Can less restrictive alternatives preserve competition?
15. Would the proposed remedy itself reduce innovation?

21. Conclusion

Strategic foresight is becoming an important analytical dimension of competition law for advanced intelligence markets.

The central challenge is that AI and related technologies can create competitive advantages through combinations of compute, data, models, talent, infrastructure, distribution, interoperability, and ecosystem effects.

Traditional antitrust analysis remains essential, but forward-looking assessment can help identify risks before market structures become irreversible.

The principal lesson from the Microsoft, Google, Meta, Android, and related cases is not that advanced-intelligence firms should automatically be treated as dominant. Rather, the cases demonstrate several recurring competition-law principles:

• control of technological bottlenecks can affect adjacent markets;
• distribution can be as important as product quality;
• vertical integration can create foreclosure opportunities;
• ecosystem arrangements can affect entry;
• nascent competitors may matter despite limited current market share; and
• interoperability and access can become important where a platform functions as a competitive bottleneck.

Accordingly, competition law and strategic foresight should focus on identifying concrete mechanisms through which present conduct or transactions could shape future market structure, while preserving incentives for genuine technological innovation.