Competition Law and Strategic Foresight for Autonomous Economy Regulation

Introduction

Strategic foresight for autonomous economy regulation refers to the use of competition-law analysis to anticipate how increasingly autonomous economic systems—AI agents, algorithmic marketplaces, autonomous vehicles, automated procurement systems, robotic production, smart contracts, digital platforms and machine-to-machine transactions—may alter competitive conditions before conventional antitrust rules are triggered.

The central competition-law concern is that autonomous systems may not merely automate existing commercial conduct. They can independently recommend prices, select trading partners, allocate resources, negotiate transactions, optimize supply chains and adapt their behaviour from data. This creates competition risks involving algorithmic collusion, self-preferencing, data concentration, interoperability barriers, exclusionary conduct, and the emergence of autonomous gatekeepers.

Strategic foresight therefore supplements traditional ex-post enforcement with an ex-ante assessment of foreseeable competitive risks.

I. Meaning of Autonomous Economy

An autonomous economy is an economic environment in which software, algorithms, AI agents, machines or connected systems perform significant commercial decisions with limited direct human intervention.

Examples include:

1. AI purchasing agents selecting suppliers.
2. Algorithmic pricing systems adjusting prices continuously.
3. Autonomous vehicles choosing routes and service providers.
4. AI-powered financial systems allocating capital.
5. Robotic factories selecting production levels.
6. Machine-to-machine commerce conducting transactions.
7. Smart-contract systems automatically executing agreements.
8. AI marketplaces matching consumers and suppliers.
9. Autonomous energy-management systems buying and selling electricity.
10. Agentic AI systems negotiating commercial terms on behalf of businesses.

Competition law must therefore examine not only agreements between humans but also competitive effects produced through technological systems.

II. Meaning of Strategic Foresight in Competition Law

Strategic foresight involves systematically examining:

• emerging technologies;
• future market structures;
• potential bottlenecks;
• technological dependencies;
• changing barriers to entry;
• data accumulation;
• interoperability;
• network effects;
• algorithmic coordination;
• possible future dominance.

It is not a prediction that a particular firm will become dominant. Rather, it is a regulatory methodology for identifying plausible competition risks early enough to permit effective intervention.

Traditional model

Conduct → Investigation → Evidence → Finding → Remedy

Strategic-foresight model

Technology → Emerging capability → Possible market structure → Competition risk → Monitoring → Early intervention

III. Why Autonomous Economies Create New Competition Problems

1. Algorithmic Coordination

Autonomous pricing systems can monitor competitors continuously.

Even without an explicit human agreement, algorithms may:

• observe competitors' prices;
• react immediately;
• converge on common prices;
• punish deviations;
• learn that aggressive competition reduces profits.

This raises the question whether conventional concepts of concerted practices are sufficiently adaptable to autonomous decision-making.

2. Algorithmic Collusion

Algorithms can potentially facilitate:

• price coordination;
• output restrictions;
• market allocation;
• bid coordination;
• customer allocation.

A major legal difficulty is determining attribution.

If two autonomous systems independently learn that maintaining high prices maximizes profit, the regulator must determine whether the resulting conduct is:

• genuinely unilateral;
• algorithmically facilitated;
• tacit coordination;
• evidence of an agreement;
• or a conventional cartel concealed through technology.

IV. Data Concentration and Autonomous Market Power

Autonomous economic systems often depend upon enormous datasets.

A firm possessing:

• transaction data;
• behavioural data;
• real-time pricing information;
• location information;
• industrial data;
• training data;

may obtain advantages that competitors cannot easily reproduce.

The resulting competitive advantage can become self-reinforcing:

More users → More data → Better AI → Better service → More users → More data

This creates a potential data-network-effect loop.

Competition law must therefore examine whether control over data becomes a source of:

• market power;
• entry barriers;
• exclusion;
• discriminatory access;
• vertical foreclosure.

V. Autonomous Gatekeepers

An autonomous platform may operate as an intermediary between several groups:

Consumers ↔ AI platform ↔ Suppliers

If the platform's AI determines:

• which suppliers appear;
• ranking;
• recommendations;
• pricing;
• access;
• advertising;
• transaction conditions,

the platform may exercise substantial control over market access.

This creates risks of algorithmic self-preferencing.

For example, an AI marketplace could systematically recommend the platform's own products over rival products even where users have not expressly requested such prioritization.

VI. Self-Preferencing by Autonomous Systems

Self-preferencing occurs when a vertically integrated platform gives preferential treatment to its own downstream products or services.

In an autonomous economy, preferential treatment may be embedded in:

• ranking algorithms;
• recommendation engines;
• search systems;
• AI assistants;
• automated procurement;
• marketplace allocation algorithms.

The difficult question becomes whether the discriminatory outcome resulted from:

1. intentional programming;
2. optimization objectives;
3. training data;
4. machine learning;
5. commercial incentives;
6. autonomous adaptation.

Competition law may therefore need to examine the design and governance of algorithms, rather than merely their final outputs.

VII. Interoperability and Autonomous Ecosystems

Autonomous systems frequently operate as ecosystems.

Examples include:

• AI operating systems;
• autonomous vehicles;
• smart-home systems;
• cloud ecosystems;
• digital payment ecosystems;
• industrial IoT systems.

A dominant undertaking may restrict interoperability by preventing rival systems from accessing:

• APIs;
• data;
• interfaces;
• identity systems;
• technical standards;
• essential software components.

The competitive concern is that interoperability restrictions may transform a temporary technological advantage into persistent ecosystem control.

VIII. Autonomous Economy and Essential Facilities

Certain autonomous markets may develop strategically important infrastructure.

Examples include:

• AI compute infrastructure;
• cloud platforms;
• digital identity infrastructure;
• payment rails;
• autonomous-vehicle charging networks;
• industrial data platforms;
• high-performance computing;
• communications networks.

Where a dominant undertaking controls infrastructure that competitors cannot reasonably reproduce, refusal of access may raise issues analogous to the essential-facilities doctrine, subject to the applicable jurisdiction's legal test.

IX. Network Effects

Autonomous markets can exhibit unusually strong network effects.

For example:

Users → Data → Better AI → Better recommendations → More users

A second effect may arise:

More suppliers → More transactions → More data → Better matching → More suppliers

These feedback loops can create rapid concentration.

Strategic foresight therefore asks:

Could today's competitive advantage become tomorrow's structural barrier to entry?

X. Autonomous Pricing

AI pricing systems may change prices in milliseconds.

They can incorporate:

• competitor prices;
• inventory;
• demand;
• consumer behaviour;
• time;
• location;
• purchasing history.

Potential competition concerns include:

A. Coordinated pricing

Several systems may converge on supra-competitive prices.

B. Discriminatory pricing

Different consumers may receive different prices.

C. Predatory pricing

An autonomous system may temporarily sacrifice margins to eliminate competitors.

D. Exclusionary pricing

An integrated platform may systematically disadvantage competing suppliers.

E. Personalized pricing

Extensive consumer data may permit highly individualized price discrimination.

XI. Autonomous Mergers and Acquisitions

Strategic foresight is particularly important for mergers involving AI firms.

A conventional turnover threshold may fail to capture the significance of a target possessing:

• valuable datasets;
• AI models;
• engineers;
• patents;
• user communities;
• computing capacity;
• strategically important algorithms.

Consequently, competition authorities increasingly need to examine innovation competition and future competitive constraints, rather than merely current revenues.

XII. Killer Acquisitions and Autonomous Economy

A dominant digital company may acquire a small AI company before the target becomes a significant competitor.

The acquisition may remove a potential future competitor.

The relevant question becomes:

Would the target have developed into an important competitive constraint absent the acquisition?

Strategic foresight therefore incorporates:

• pipeline products;
• innovation capabilities;
• R&D trajectories;
• technological complementarities;
• future market entry.

XIII. Autonomous Economy and Innovation Competition

Competition does not occur solely through current prices.

Firms may compete through:

• better AI models;
• improved robotics;
• autonomous decision-making;
• privacy;
• security;
• accuracy;
• interoperability;
• energy efficiency.

A transaction or exclusionary practice that eliminates a significant innovation pathway may harm competition even when short-term prices do not increase.

XIV. Six Important Case Laws

1. United States v. Airline Tariff Publishing Co. (1994)

The U.S. Department of Justice challenged airline practices involving computerized fare dissemination.

Competition significance

The case demonstrated how computerized systems can facilitate coordination between competitors.

The important foresight lesson is that technology does not need to contain an explicit cartel agreement to become commercially relevant to coordination.

Relevance to autonomous economy

Modern AI systems can perform the same monitoring and reaction functions at much greater speed.

Therefore:

Computerized information exchange → Algorithmic monitoring → Potential coordination

2. United States v. Topkins (2015)

This case involved an online poster retailer using algorithms to coordinate prices for posters sold through Amazon Marketplace.

Topkins pleaded guilty to participating in a conspiracy involving algorithmic pricing.

Competition significance

The case is particularly important because software was used as an instrument for implementing a traditional price-fixing arrangement.

Foresight lesson

A firm cannot escape cartel liability merely because its employees implement the agreement through software.

The relevant legal principle is:

Technology can be the mechanism through which anticompetitive conduct is implemented.

3. Eturas UAB v. Lietuvos Respublikos konkurencijos taryba (CJEU, 2016)

This case concerned an online travel-booking platform and a system message that imposed a restriction affecting discounts offered by travel agencies.

The Court of Justice examined when conduct facilitated through an electronic platform could support an inference of participation in a concerted practice.

Competition significance

The case illustrates the evidentiary importance of:

• platform communications;
• electronic systems;
• knowledge;
• participation;
• subsequent conduct.

Autonomous-economy relevance

It demonstrates that competition law can attach significance to digital architecture and electronic communications, not merely traditional meetings and written contracts.

4. Google Shopping — Google and Alphabet v Commission (CJEU, 2024)

The EU litigation concerning Google's comparison-shopping practices addressed preferential treatment of Google's own comparison-shopping service in general search results.

Competition significance

The case is central to understanding digital self-preferencing and the use of a dominant platform's infrastructure to favour its own service.

Autonomous-economy relevance

AI assistants increasingly determine what information, products or services users see.

The same structural issue can therefore arise where an autonomous recommendation system controls commercial visibility.

5. Google Android — Google and Alphabet v Commission (General Court, 2022)

The European Commission's Android case concerned contractual restrictions involving Google's Android ecosystem, including tying and restrictions affecting competing search services.

The General Court substantially upheld the Commission's infringement decision while adjusting certain aspects of the reasoning and penalty.

Competition significance

The case illustrates how an ecosystem owner can use control over important technological layers to influence adjacent markets.

Autonomous-economy relevance

The case is relevant to future AI ecosystems where one company may control:

Operating system → AI assistant → App distribution → Search → Data → Advertising

Strategic foresight therefore requires regulators to examine ecosystem leverage, not just isolated products.

6. United States v. Google LLC — Search and Search Advertising (D.D.C., 2024)

The U.S. federal litigation concerning Google's search distribution practices addressed agreements through which Google obtained default or distribution positions that reinforced its position in general search.

Competition significance

The litigation illustrates the importance of distribution advantages, defaults and network effects in digital markets.

Autonomous-economy relevance

AI assistants may increasingly become the gateway through which consumers access information and commercial services.

Control over the default AI interface could consequently become an important competitive bottleneck.

XV. Additional Relevant Authorities

Several other decisions provide useful principles for autonomous-economy regulation.

7. United States v. Apple Inc. (2024)

The U.S. government's antitrust case against Apple raises issues concerning ecosystem restrictions, interoperability and exclusionary conduct.

It is relevant to autonomous ecosystems because control over a tightly integrated technological environment can influence competition in adjacent markets.

8. Qualcomm Inc. v. FTC (9th Cir., 2020)

The litigation concerned Qualcomm's licensing practices and the competitive effects of its conduct in cellular technology markets.

Its broader significance lies in examining how control over technologically important intellectual property and licensing arrangements can affect downstream competition.

9. Intel v Commission (CJEU, 2017)

The Intel litigation concerned loyalty rebates and exclusionary effects.

The judgment is relevant to autonomous markets because technological markets may contain substantial fixed costs and strong customer dependencies, making foreclosure analysis particularly important.

10. Microsoft Corp. v Commission (General Court, 2007)

Microsoft's conduct concerning interoperability and tying remains important to understanding competition in technology ecosystems.

The case demonstrates how control over one technological layer can potentially restrict competition in adjacent markets.

XVI. Strategic Foresight Framework

Competition authorities dealing with autonomous economies can use a five-stage framework.

Stage 1 — Technology Mapping

Identify:

• AI systems;
• autonomous agents;
• APIs;
• data infrastructures;
• cloud dependencies;
• computing resources;
• algorithms.

↓

Stage 2 — Market Mapping

Identify:

• suppliers;
• consumers;
• intermediaries;
• competitors;
• potential entrants;
• infrastructure providers.

↓

Stage 3 — Dependency Mapping

Determine whether firms depend upon:

• data;
• cloud computing;
• operating systems;
• AI models;
• payment infrastructure;
• APIs;
• distribution platforms.

↓

Stage 4 — Competition-Risk Mapping

Examine:

• collusion;
• exclusion;
• self-preferencing;
• tying;
• refusal to deal;
• discriminatory access;
• predatory pricing;
• excessive pricing;
• interoperability restrictions;
• acquisitions of emerging competitors.

↓

Stage 5 — Regulatory Response

Possible responses include:

• monitoring;
• behavioural commitments;
• interoperability requirements;
• data-access obligations;
• merger review;
• structural remedies;
• algorithmic auditing;
• transparency requirements;
• conventional antitrust enforcement.

XVII. Algorithmic Governance as a Competition-Law Issue

Autonomous systems require governance mechanisms.

Competition authorities may need to examine:

1. Objective functions

What commercial objective has the algorithm been designed to maximize?

2. Training data

Does the dataset systematically favour one undertaking?

3. Feedback mechanisms

Does the system learn from competitors' behaviour?

4. Access rules

Who can use the system?

5. Interoperability

Can competing systems connect?

6. Auditability

Can regulators reconstruct how an important commercial decision was made?

XVIII. Ex-Ante and Ex-Post Competition Regulation

The two approaches should complement rather than replace one another.

XIX. Regulatory Challenges

1. Attribution

Who is responsible for autonomous conduct?

• developer;
• platform operator;
• deploying business;
• user;
• owner of the model?

2. Explainability

Competition authorities may need to understand why an algorithm produced a particular outcome.

However, machine-learning models can be difficult to interpret.

3. Dynamic Markets

Traditional market definition may become difficult where technology changes rapidly.

4. Multi-Market Effects

A single AI system may simultaneously operate across:

• search;
• advertising;
• commerce;
• finance;
• logistics;
• cloud;
• communications.

Competition analysis may therefore require an ecosystem perspective.

5. Regulatory Lag

Technological development may occur faster than legislative reform.

Strategic foresight attempts to reduce this regulatory gap.

XX. Remedies for Autonomous-Economy Competition Problems

Possible remedies include:

Behavioural remedies

• prohibiting discriminatory ranking;
• restricting data use;
• preventing algorithmic coordination;
• requiring non-discriminatory access.

Structural remedies

• separation of business units;
• divestiture;
• restrictions on acquisitions.

Technical remedies

• API access;
• interoperability;
• portability;
• technical standards.

Governance remedies

• algorithmic auditing;
• logging;
• independent monitoring;
• compliance systems.

XXI. Key Doctrinal Principles

The following competition-law principles are particularly important for autonomous economies:

1. Technology does not immunize anticompetitive conduct.
2. Algorithmic implementation can evidence or facilitate traditional antitrust violations.
3. Data can constitute an important competitive input.
4. Interoperability can be critical to maintaining competitive constraints.
5. Platform architecture can affect market access.
6. Autonomous pricing requires careful analysis of coordination and unilateral conduct.
7. AI acquisitions can raise innovation-competition concerns.
8. Network effects can rapidly reinforce market power.
9. Competition authorities may need to investigate ecosystem-level dependencies.
10. Strategic foresight should complement—not replace—evidence-based enforcement.

Conclusion

Strategic foresight for autonomous economy regulation represents an important evolution in competition-law thinking. Autonomous economic systems can transform the way markets coordinate, compete and allocate resources. The principal challenge is that traditional antitrust analysis frequently examines conduct after it has occurred, whereas autonomous markets can develop feedback loops and technological dependencies extremely quickly.

The cases involving Airline Tariff Publishing, Topkins, Eturas, Google Shopping, Google Android and Google's search-distribution practices demonstrate different ways in which technology, digital platforms, algorithms, defaults and ecosystem control can interact with competition law.

The future regulatory challenge is therefore not simply whether AI or autonomous systems should be regulated, but whether competition law can identify when autonomous systems:

increase competitive efficiency → facilitate coordination → create exclusionary dependencies → entrench market power.

Strategic foresight provides a framework for identifying that transition while preserving the central competition-law objectives of competitive markets, innovation, consumer choice, market access and contestability.