Competition Law and Competition Governance in Autonomous Civilizations

1. Introduction

“Autonomous civilizations” is a forward-looking concept describing societies in which AI systems, autonomous agents, algorithmic institutions, robotics, decentralized networks, smart infrastructure, and machine-to-machine commerce perform functions traditionally controlled by human decision-makers. Examples could include autonomous procurement, algorithmic allocation of resources, self-executing contracts, AI-managed logistics, automated financial markets, autonomous mobility, and decentralized digital ecosystems.

Competition law in such an environment cannot focus only on conventional questions such as whether two human-controlled companies have agreed to fix prices. The central issue becomes:

How should competition be preserved when economically significant decisions are increasingly made by autonomous systems that may learn, coordinate, optimize, or exclude without direct human instructions?

Existing competition law nevertheless provides several useful foundations: prohibitions on cartels, abuse of dominance, merger control, essential-facility principles, discriminatory access, tying, interoperability, and restrictions on technological ecosystems.

2. Meaning of Competition Governance

Competition law establishes enforceable legal rules against anticompetitive conduct.

Competition governance is broader. It includes:

• competition legislation;
• regulatory institutions;
• market-design rules;
• algorithmic auditing;
• interoperability requirements;
• data-access rules;
• transparency obligations;
• merger supervision;
• technical standards;
• monitoring of autonomous agents;
• remedies for algorithmic exclusion;
• accountability of firms deploying autonomous systems.

Thus:

Competition Law = rules against anticompetitive conduct

Competition Governance = institutional architecture for maintaining competitive markets

In autonomous civilizations, governance becomes particularly important because traditional ex-post enforcement may occur only after an autonomous ecosystem has already become difficult to challenge.

3. Why Autonomous Civilizations Create Competition Problems

A. Autonomous decision-making

An AI agent may independently determine:

• prices;
• suppliers;
• customers;
• advertising expenditure;
• inventory;
• credit;
• procurement;
• access conditions;
• delivery routes;
• platform rankings.

This creates a question of attribution.

If an autonomous system produces an exclusionary outcome, competition authorities must determine whether liability attaches to:

1. the company;
2. the developers;
3. the operator;
4. the users of the system;
5. several participants jointly.

Competition law generally regulates economic actors rather than algorithms as independent legal persons.

B. Algorithmic coordination

Autonomous systems may observe competitors' prices and continuously optimize their own strategies.

This creates several possibilities:

• explicit human coordination;
• algorithmic implementation of an agreement;
• conscious parallelism;
• hub-and-spoke coordination;
• tacit algorithmic coordination;
• autonomous convergence without human communication.

The difficult question is whether competition law should intervene only where there is proof of an agreement or whether certain forms of predictable algorithmic coordination require additional governance mechanisms.

4. Relevant Competition-Law Framework

Autonomous markets can be analysed principally through five areas.

1. Anti-cartel rules

These address:

• price fixing;
• market allocation;
• output restriction;
• bid rigging;
• information exchange.

2. Abuse of dominance

Potential autonomous-system abuses include:

• self-preferencing;
• discriminatory algorithms;
• exclusionary ranking;
• refusal of interoperability;
• predatory algorithmic pricing;
• tying;
• data foreclosure.

3. Merger control

Autonomous ecosystems increase the importance of acquisitions involving:

• AI startups;
• data companies;
• foundation models;
• cloud infrastructure;
• robotics;
• semiconductor technologies;
• digital platforms.

4. Essential facilities and access

A dominant autonomous infrastructure may become indispensable where competitors depend upon:

• cloud computing;
• AI models;
• payment infrastructure;
• identity systems;
• data repositories;
• interoperability protocols;
• communications networks.

5. Digital-platform governance

Competition authorities may need to examine:

• gatekeeper power;
• interoperability;
• data portability;
• switching costs;
• ecosystem lock-in;
• default settings;
• app-store restrictions;
• algorithmic ranking.

5. Autonomous Civilizations and Market Power

Traditional market power is often associated with price.

Autonomous markets require a broader conception.

Relevant indicators may include:

• control over computational infrastructure;
• control over datasets;
• control over AI models;
• control over autonomous agents;
• control over standards;
• control over interfaces;
• control over identity systems;
• switching costs;
• network effects;
• ecosystem dependence;
• access to physical infrastructure.

Consequently, a company could possess substantial competitive power even when its service is offered at a zero monetary price.

6. Competition Governance of Autonomous Agents

A future governance framework could impose obligations on companies deploying economically significant autonomous systems.

Possible obligations

A. Auditability

Systems should preserve sufficient records to reconstruct economically significant decisions.

B. Explainability

Where an automated decision materially excludes a competitor, regulators may need access to the relevant decision logic.

C. Non-discrimination

Autonomous systems controlling infrastructure should not systematically discriminate between similarly situated competitors without objective justification.

D. Interoperability

Dominant platforms may be required in appropriate circumstances to permit technical interoperability.

E. Data portability

Users and businesses should be capable of moving relevant data between competing systems.

F. Algorithmic compliance

Companies should conduct competition-law risk assessments before deploying autonomous pricing or allocation systems.

7. Important Case Laws

Because “autonomous civilizations” is a future-oriented concept, there are no reported cases specifically concerning an autonomous civilization. The following cases provide established competition-law principles that can be applied to autonomous and algorithmically governed markets.

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

The U.S. Department of Justice's case against Apple concerns alleged exclusionary conduct relating to the iPhone ecosystem.

Relevance

The case illustrates how control over an integrated technological ecosystem can generate competition concerns involving:

• interoperability;
• access restrictions;
• platform dependence;
• ecosystem control;
• barriers to competing products.

For autonomous ecosystems, the same principle becomes relevant where one operator controls the interfaces through which autonomous agents interact.

Principle

Competition analysis can extend beyond the price of a product to the architecture and conduct through which an ecosystem protects its competitive position.

2. European Commission v. Google (Google Shopping)

The European Commission found that Google had abused its dominant position by giving preferential treatment to its own comparison-shopping service in search results.

Relevance

This is highly relevant to autonomous civilizations because AI systems may increasingly determine:

• which products are displayed;
• which suppliers are selected;
• which services receive priority;
• which autonomous agents obtain access.

Principle

A dominant intermediary's control over ranking and visibility can create competition concerns when the system systematically advantages its own downstream service.

3. Google Android — European Commission

The European Commission's Android decision concerned Google's contractual arrangements involving Android devices, including restrictions relating to search and app distribution.

Relevance

Autonomous ecosystems may similarly use:

• default settings;
• technical integration;
• bundling;
• contractual restrictions;
• ecosystem dependencies.

Principle

Control of an operating environment can affect competition in adjacent markets where rivals depend upon access to that environment.

4. Google AdSense — European Commission

The European Commission found competition concerns concerning contractual restrictions imposed by Google in online advertising intermediation.

Relevance

Autonomous advertising systems may automatically select:

• advertising inventory;
• advertisers;
• prices;
• targeting mechanisms;
• competing exchanges.

If the operator of the infrastructure also competes downstream, algorithmic control can create conflicts of interest.

Principle

Vertical contractual restrictions may become problematic when they reinforce the position of a dominant intermediary and restrict rival access.

5. United States v. Microsoft Corp. (D.C. Cir. 2001)

The Microsoft litigation concerned Microsoft's conduct involving the Windows operating-system platform and competing technologies.

Relevance

It remains particularly important for autonomous civilizations because operating systems may evolve into general-purpose environments for autonomous agents.

A dominant platform may control:

• APIs;
• defaults;
• application access;
• interoperability;
• technical standards.

Principle

A dominant platform can attract competition-law scrutiny where its control over an important technological platform is used to disadvantage competing products.

6. United States v. Terminal Railroad Association (1912)

This classic U.S. Supreme Court case concerned control over essential railroad infrastructure and discriminatory access.

Relevance

Its underlying principle is particularly useful for autonomous civilization scenarios involving:

• shared digital infrastructure;
• autonomous transportation networks;
• AI compute infrastructure;
• communications systems;
• common data networks.

Principle

Control over infrastructure indispensable to effective competition may create competition concerns where access is denied or discriminatory.

7. MCI Communications Corp. v. AT&T (1983)

The U.S. Court of Appeals for the Seventh Circuit developed an influential framework for refusal-to-deal and essential-facility analysis.

Relevance

Autonomous markets may create new essential facilities such as:

• cloud infrastructure;
• AI compute;
• digital identity;
• autonomous-network interfaces;
• proprietary technical protocols.

Principle

A dominant firm controlling an indispensable facility can face competition-law scrutiny when access is withheld under circumstances that substantially impair competition.

8. United Brands v Commission (1978)

The Court of Justice of the European Union examined abuse of dominance and discriminatory commercial conditions.

Relevance

Autonomous platforms could potentially use automated systems to discriminate among:

• suppliers;
• customers;
• geographic markets;
• competing autonomous agents.

Principle

Dominant firms have special responsibilities not to use their market power in ways that distort competitive conditions.

9. Hoffmann-La Roche v Commission (1979)

The CJEU established important principles concerning exclusionary practices and loyalty-inducing arrangements by dominant firms.

Relevance

Autonomous ecosystems could produce similar effects through automated:

• loyalty incentives;
• exclusivity;
• preferential access;
• bundling;
• resource allocation.

Principle

Conduct that strengthens dominance by restricting customers' ability to deal with competitors may constitute an abuse depending upon its circumstances and effects.

8. Algorithmic Pricing and Autonomous Cartels

One of the most difficult questions concerns algorithmic collusion.

Imagine four autonomous logistics systems:

1. System A monitors market prices.
2. System B does the same.
3. Each algorithm is designed to maximize profit.
4. Each rapidly responds to the other's pricing.
5. Prices converge and remain elevated.

There may be no conventional meeting or telephone call.

Competition law therefore needs to distinguish:

Type I — Explicit human cartel

Humans agree and algorithms implement the agreement.

Traditional cartel rules apply strongly.

Type II — Algorithmic implementation

Humans communicate indirectly through algorithms designed to coordinate.

Evidence of communication and intent becomes important.

Type III — Independent algorithmic adaptation

Each system independently responds to public information.

This raises a more difficult question because parallel conduct alone does not necessarily establish an unlawful agreement.

Type IV — Autonomous coordination

Algorithms independently develop strategies that produce stable coordination.

This creates a potential gap between traditional agreement-based competition law and technologically generated market outcomes.

9. Autonomous Procurement Markets

Autonomous procurement could fundamentally change public and private purchasing.

An AI system might automatically:

• identify suppliers;
• solicit bids;
• evaluate bids;
• negotiate;
• select suppliers;
• renew contracts.

Competition governance should address:

• bid-rigging;
• algorithmic supplier exclusion;
• discriminatory scoring;
• automated collusion;
• opaque procurement criteria;
• preferential treatment of affiliated suppliers.

The procurement algorithm itself may become a market-design mechanism.

Therefore, competition compliance must occur at the design stage rather than solely after an infringement.

10. Autonomous Data Markets

Data may function as a critical competitive input.

Competition issues can arise where a dominant entity controls:

• consumer data;
• industrial data;
• mobility data;
• transaction data;
• behavioural information;
• training datasets.

Potential concerns include:

Data foreclosure

A dominant company prevents competitors from obtaining necessary data.

Data tying

Access to one service is conditioned on surrendering or purchasing another data-related service.

Data discrimination

Competitors receive inferior data access.

Data accumulation

Acquisitions permit a dominant firm to combine datasets in ways that substantially increase entry barriers.

11. Autonomous Infrastructure

Future infrastructure could be managed largely by intelligent systems.

Examples include:

• autonomous electricity grids;
• smart transportation systems;
• AI-managed telecommunications;
• robotic logistics networks;
• autonomous ports;
• intelligent water systems;
• automated financial infrastructure.

Competition governance should ensure that infrastructure operators do not use control of essential infrastructure to eliminate downstream competition.

This revives the logic of essential facilities, access regulation and non-discrimination.

12. Autonomous Mergers and Acquisitions

Merger control becomes especially important because autonomous ecosystems may exhibit strong network effects.

A seemingly small acquisition could provide a dominant platform with:

• a critical AI model;
• proprietary data;
• a key interface;
• an autonomous-agent protocol;
• a cybersecurity technology;
• an emerging competitor.

Therefore, conventional turnover-based thresholds may sometimes fail to capture economically significant acquisitions.

Authorities may need to examine:

• transaction value;
• innovation potential;
• data assets;
• future competitive significance;
• interoperability;
• ecosystem effects;
• nascent competition.

13. The Role of Human Accountability

Autonomy should not automatically eliminate legal responsibility.

A useful governance principle is:

Autonomous operation does not necessarily mean autonomous legal responsibility.

The enterprise that designs, deploys, supervises, benefits from, or controls an autonomous system may remain responsible for competition-law compliance.

This is comparable to the broader principle that corporations cannot ordinarily avoid regulatory obligations merely because decisions are technologically mediated.

14. Ex-Ante and Ex-Post Governance

Ex-post enforcement

Authorities investigate after conduct occurs.

Advantages:

• respects market freedom;
• focuses on actual harm;
• avoids unnecessary intervention.

Disadvantages:

• autonomous markets may evolve extremely rapidly;
• network effects may make reversal difficult;
• exclusion may become entrenched.

Ex-ante governance

Rules are imposed before harmful conduct occurs.

Possible measures include:

• interoperability;
• data portability;
• algorithmic audits;
• transparency;
• access obligations;
• merger notification;
• non-discrimination requirements.

The future challenge is determining which markets require ex-ante regulation and which should remain primarily subject to ordinary competition enforcement.

15. Competition Governance Model for Autonomous Civilizations

A possible framework can be represented as:

Autonomous Technology

↓

Market Identification

↓

Market Power Assessment

↓

Algorithmic Conduct Assessment

↓

Data + Infrastructure Access Analysis

↓

Interoperability Assessment

↓

Merger / Acquisition Review

↓

Algorithmic Compliance Audit

↓

Ex-Ante / Ex-Post Remedy

↓

Continuous Competition Monitoring

16. Key Governance Principles

Principle 1 — Technological neutrality

Competition law should regulate competitive effects rather than simply regulate particular technologies.

Principle 2 — Human accountability

Autonomous systems should not become mechanisms for avoiding corporate responsibility.

Principle 3 — Interoperability

Where justified by market conditions, interoperability can reduce ecosystem foreclosure.

Principle 4 — Contestability

Markets should remain open to entry and innovation.

Principle 5 — Algorithmic neutrality

Dominant systems should not systematically favour affiliated businesses without legitimate justification.

Principle 6 — Data access

Control over indispensable data should be assessed as a potential source of market power.

Principle 7 — Auditability

Regulators should have sufficient information to reconstruct significant automated decisions.

Principle 8 — Proportionality

Not every autonomous optimization is an antitrust violation. Intervention should correspond to the competitive harm and relevant legal standard.

17. Major Future Competition Risks

18. Central Legal Challenge

The fundamental difficulty is that traditional competition law assumes that economic decisions are ultimately attributable to identifiable human or corporate actors.

Autonomous civilizations introduce systems where:

The market itself increasingly becomes algorithmically organized.

This changes competition analysis from merely asking:

“Did the company engage in anticompetitive conduct?”

to a broader set of questions:

1. Who controls the autonomous system?
2. Who owns the underlying infrastructure?
3. Who controls the data?
4. Can competitors access the system?
5. Can users switch?
6. Can rival autonomous agents interoperate?
7. Can the system coordinate with competing systems?
8. Can regulators audit economically significant decisions?
9. Can new competitors enter the ecosystem?
10. Can competition be restored after exclusion occurs?

19. Conclusion

Competition law in autonomous civilizations will likely evolve from a system primarily concerned with human agreements and corporate conduct into a broader framework concerned with market architecture, autonomous decision-making, data, infrastructure, interoperability and algorithmic governance.

The traditional doctrines remain important. Microsoft, United Brands, Hoffmann-La Roche, Google Shopping, Google Android, Terminal Railroad and MCI demonstrate that competition law already possesses principles capable of addressing platform control, exclusion, discriminatory access, technological integration and essential infrastructure.

The principal future challenge is not whether autonomous systems should be subject to competition law, but how responsibility, market power and competitive effects should be attributed when economically significant decisions are increasingly made by autonomous technological systems.