Competition Law and Competition Governance in Post-Human Economic Systems

1. Introduction

A post-human economic system is a useful conceptual term for markets in which economic activity is increasingly organized not only by human individuals and firms, but also by AI systems, autonomous agents, algorithms, robots, digital twins, biometric systems, machine-learning models, autonomous platforms, decentralized networks and other computational actors.

The expression does not mean that humans disappear from the economy. Rather, it describes an economic environment in which machines increasingly make, recommend, optimize, negotiate or execute decisions that traditionally belonged to human market participants.

This creates a fundamental competition-law question:

How should competition law operate when economically significant decisions are increasingly made by systems that are autonomous, adaptive, data-driven and capable of interacting with one another without direct human intervention?

Traditional competition law is principally concerned with:

• agreements between undertakings;
• abuse of dominant position;
• mergers and acquisitions;
• exclusionary conduct;
• collusion;
• consumer welfare;
• market access; and
• competitive effects.

Post-human markets add further problems:

• algorithmic coordination;
• autonomous pricing;
• AI-controlled allocation;
• data monopolisation;
• machine-to-machine contracting;
• AI ecosystem lock-in;
• autonomous agents acting on behalf of consumers;
• interoperability between intelligent systems;
• control of computational infrastructure;
• access to training data and models;
• self-preferencing by algorithmic platforms;
• and competition between human and machine decision-makers.

Recent enforcement demonstrates that competition authorities are already dealing with several of these issues. For example, the European Commission's Digital Markets Act enforcement has addressed self-preferencing, steering and interoperability, while U.S. authorities have pursued algorithmic coordination theories in the RealPage litigation.

2. Meaning of Competition Governance in a Post-Human Economy

Competition governance is broader than conventional antitrust enforcement.

It involves the institutional rules and mechanisms through which competitive conditions are maintained in markets increasingly mediated by intelligent technologies.

It may include:

1. competition legislation;
2. sectoral regulation;
3. digital-platform regulation;
4. interoperability obligations;
5. data-access requirements;
6. algorithmic auditing;
7. merger control;
8. regulatory sandboxes;
9. technical standards;
10. transparency obligations;
11. remedies against ecosystem lock-in;
12. monitoring of automated decision-making; and
13. international regulatory cooperation.

Thus:

Competition law = legal prohibition and correction of anticompetitive conduct

while

Competition governance = the broader architecture for preserving contestability, openness and competitive neutrality.

3. Characteristics of Post-Human Economic Systems

A. Autonomous economic decision-making

AI systems can determine:

• prices;
• inventory;
• advertising;
• credit allocation;
• procurement;
• routing;
• investment;
• resource allocation;
• and consumer recommendations.

Consequently, the traditional distinction between a human decision-maker and an automated tool becomes less meaningful.

B. Machine-to-machine interaction

Two firms may deploy algorithms that continuously interact.

For example:

Algorithm A → observes market → changes price → Algorithm B responds → Algorithm A learns → price stabilisation

No executive may expressly communicate with the competitor.

This raises the question whether competition law should intervene where coordination emerges from algorithmic interaction rather than a conventional human agreement.

The RealPage litigation is particularly important because the DOJ alleged that competing landlords supplied non-public information to a common pricing system and received algorithmically generated pricing recommendations.

4. Market Definition in Post-Human Markets

Traditional market definition may become difficult because an intelligent ecosystem can simultaneously operate across several markets.

For example, an AI platform may provide:

• search;
• advertising;
• cloud computing;
• payments;
• identity;
• AI models;
• data analytics;
• consumer assistants.

A conventional single-product market may therefore fail to capture the competitive structure.

Important concepts include:

1. Ecosystem markets

Competition may occur between ecosystems rather than individual products.

2. Multi-sided markets

An AI platform may simultaneously connect:

• consumers;
• advertisers;
• developers;
• merchants;
• data suppliers; and
• AI agents.

3. Zero-price markets

The absence of monetary pricing does not necessarily mean the absence of competitive harm.

Competition may instead occur through:

• data;
• attention;
• quality;
• privacy;
• interoperability;
• innovation;
• and computational resources.

5. Data as a Competitive Resource

In post-human economies, data can become a fundamental competitive input.

An incumbent may possess:

• user data;
• behavioural data;
• transaction histories;
• training datasets;
• search queries;
• location information;
• biometric information;
• industrial datasets;
• machine-generated data.

A dominant firm can potentially use these datasets to improve its AI systems, thereby creating a feedback loop:

More users → more data → better model → better service → more users → more data

This can produce data-driven network effects.

Competition governance may therefore need to address:

• discriminatory access to data;
• exclusive data arrangements;
• data portability;
• data interoperability;
• refusal to provide essential datasets;
• combining datasets across markets;
• and discriminatory use of third-party data.

6. Algorithmic Collusion

Algorithmic systems create a particularly important competition problem.

Traditional cartel theory generally looks for:

• communication;
• agreement;
• coordination;
• concerted practices;
• exchange of commercially sensitive information.

In an algorithmic market, coordination may occur through:

• common pricing software;
• shared data;
• machine-learning systems;
• automated monitoring;
• algorithmic signalling;
• or repeated autonomous interaction.

RealPage

The DOJ's RealPage proceedings provide a contemporary example. The government alleged that competing landlords supplied competitively sensitive information to RealPage's pricing system and used its recommendations in setting rents. The DOJ subsequently proposed settlements containing restrictions on the use of competitors' non-public information and certain algorithmic pricing practices.

The case demonstrates that automation does not necessarily remove human firms from antitrust responsibility.

7. Self-Preferencing by Intelligent Systems

A post-human platform may control the algorithm determining which products consumers see.

Suppose an AI marketplace ranks:

1. its own products;
2. its own logistics service;
3. its own payment system;

above competing services.

The AI is therefore not merely processing information—it is structuring the competitive environment.

Google Shopping

In Google and Alphabet v Commission (Google Shopping), Case T-612/17, the EU General Court largely upheld the Commission's finding that Google abused its dominant position by favouring its own comparison-shopping service over competing comparison-shopping services.

This principle has major relevance to post-human systems because AI ranking systems may become the new gateways through which consumers discover products.

8. Six Important Case Laws

Case 1: Google Shopping

Google and Alphabet v European Commission, T-612/17 (General Court, 2021)

Principle

The case concerned Google's treatment of its own comparison-shopping service in search results.

The General Court largely upheld the Commission's decision and the €2.42 billion fine.

Relevance to post-human systems

The case demonstrates that:

• ranking algorithms can affect competition;
• visibility is itself an important competitive resource;
• control over an information gateway can produce exclusionary effects;
• algorithmic neutrality can become a competition-law issue.

Post-human application

An AI shopping assistant could potentially:

• rank its owner's products first;
• suppress competing products;
• manipulate recommendations;
• optimise results toward affiliated businesses.

Competition governance may therefore require algorithmic non-discrimination and transparent ranking rules.

Case 2: Google Android

Google and Alphabet v European Commission (Google Android), T-604/18 (General Court, 2022)

The General Court considered Google's conduct concerning Android, Google Search, Chrome, Play Store and related agreements.

The case involved product bundling, exclusivity payments and anti-fragmentation obligations, with the Court considering the operation of Google's mobile ecosystem.

Relevance

This case demonstrates the importance of:

• ecosystem power;
• tying;
• platform control;
• default settings;
• network effects;
• and exclusionary restrictions.

Post-human application

Future AI ecosystems may combine:

AI model + operating system + assistant + cloud + search + payments + identity

If a dominant ecosystem makes interoperability conditional upon accepting its own services, competition can be weakened.

Case 3: Amazon Marketplace

Amazon Marketplace — European Commission proceedings, 2022

The European Commission investigated Amazon's use of non-public marketplace seller data and concerns relating to the Buy Box and Prime.

Amazon offered commitments concerning:

• use of non-public seller data;
• unbiased Buy Box selection;
• treatment of marketplace sellers;
• logistics choices; and
• carrier access. 

Relevance

This illustrates a particularly important post-human problem:

The platform may simultaneously operate the marketplace and compete against the participants whose data it controls.

Post-human application

An AI marketplace operator could observe:

• competitor prices;
• demand;
• inventory;
• conversion rates;
• consumer behaviour;

and use those datasets to improve its own products.

This creates a potential information asymmetry between platform and dependent competitors.

Case 4: Meta / Facebook Marketplace

Meta Platforms – Facebook Marketplace, European Commission, 2024

The European Commission found that Meta abused a dominant position by tying Facebook Marketplace to Facebook and imposing unfair trading conditions on competing online classified-advertising services. The Commission imposed a €797.72 million fine.

Competition-law significance

The case demonstrates the importance of:

• tying;
• platform ecosystems;
• leveraging dominance from one service into another;
• unfair trading conditions.

Post-human significance

An AI ecosystem may similarly combine:

identity + social graph + AI assistant + marketplace + payment + advertising.

The larger the ecosystem, the greater the possibility of leveraging dominance from one layer into another.

Case 5: Epic Games v Apple

Epic Games, Inc. v Apple Inc., 559 F. Supp. 3d 898 (N.D. Cal. 2021), aff'd in part and rev'd in part, 67 F.4th 946 (9th Cir. 2023)

The litigation concerned Apple's App Store ecosystem, payment restrictions and relationships between developers and Apple's platform.

The U.S. DOJ continues to identify the litigation as an important antitrust matter.

Relevance

The case illustrates competition concerns surrounding:

• app-store control;
• payment systems;
• platform rules;
• alternative distribution;
• commissions;
• developer access.

Post-human application

An AI ecosystem could become a distribution gatekeeper for autonomous agents.

For example:

Consumer → AI agent → AI marketplace → application → payment

If one company controls the AI-agent marketplace, it may acquire gatekeeper power over thousands of downstream services.

Case 6: RealPage Algorithmic Pricing Litigation

United States et al. v RealPage, Inc.

The DOJ alleged that RealPage's pricing system used non-public, competitively sensitive information supplied by competing landlords and generated pricing recommendations that could reduce independent price competition.

The DOJ's later proposed settlement would restrict the use of competitors' non-public information in real-time pricing and impose limits on certain model-training practices.

Post-human significance

This is particularly important because it directly raises the problem of:

Can competition law apply when economically significant coordination is mediated by an algorithm?

The answer emerging from enforcement practice is that the presence of an algorithm does not immunise the underlying conduct from antitrust scrutiny.

9. AI Agents as Economic Actors

One of the most important developments in post-human competition is the rise of AI agents acting on behalf of consumers or businesses.

An AI agent could:

• search for products;
• compare prices;
• negotiate;
• purchase goods;
• switch suppliers;
• manage investments;
• select insurance;
• arrange logistics;
• negotiate contracts.

This creates a new market structure:

Human consumer → AI agent → platform → supplier

The AI agent becomes an intermediary between consumer and seller.

Competition law will have to consider whether the agent:

• independently chooses among suppliers;
• is controlled by a dominant platform;
• receives preferential payments;
• has access to exclusive data;
• steers users toward affiliated suppliers;
• or discriminates against competing providers.

10. AI Interoperability

Interoperability is likely to become one of the central principles of post-human competition governance.

If AI Agent A cannot interact with:

• Agent B;
• payment system C;
• cloud D;
• identity service E;

the owner of one ecosystem may create substantial switching costs.

The EU's current DMA enforcement illustrates the direction of travel: in July 2026, the European Commission issued binding specification measures concerning interoperability between competing AI services and Android features, as well as access by third-party search engines to Google Search data.

This indicates that technical interoperability can become a competition remedy, not merely an engineering issue.

11. Essential Facilities in Post-Human Markets

Traditional essential-facility concepts may acquire new importance.

Potentially critical resources include:

• AI compute;
• cloud infrastructure;
• foundation models;
• semiconductor capacity;
• large datasets;
• digital identity systems;
• payment rails;
• app stores;
• autonomous-vehicle infrastructure;
• telecommunications networks.

A dominant undertaking controlling a critical input could potentially exclude downstream competitors.

The competition-law question becomes:

When does control over computational infrastructure become sufficiently important to justify mandated access?

This requires careful assessment because forced access can also reduce incentives to invest.

12. Network Effects

Post-human markets may experience exceptionally strong network effects.

For example:

More users

↓

More data

↓

Better AI model

↓

Better predictions

↓

More users

↓

More data

This creates a data-network-effect feedback loop.

Competition authorities may therefore need to examine:

• data accumulation;
• switching costs;
• interoperability;
• multi-homing;
• portability;
• exclusivity;
• acquisition of emerging competitors.

13. Algorithmic Mergers

Traditional merger analysis often asks whether two firms' combination will substantially lessen competition.

In post-human markets, authorities may additionally examine:

Data concentration

Will the merger combine datasets that competitors cannot replicate?

Model concentration

Will the merged company control important AI models?

Compute concentration

Will the transaction increase control over scarce computing resources?

Talent concentration

Will acquisition eliminate an emerging technological rival?

Ecosystem concentration

Will the transaction integrate previously competing layers?

Future competition

Could a small AI company become a significant future competitor?

Thus, merger control must increasingly consider innovation competition and potential competition, not merely current market shares.

14. Competition and Algorithmic Discrimination

Algorithms can discriminate between:

• consumers;
• suppliers;
• merchants;
• developers;
• advertisers;
• geographic areas.

Examples include:

• different prices;
• different search rankings;
• different access conditions;
• different commission rates;
• different API access;
• different visibility.

Competition law should distinguish between:

legitimate personalised optimisation

and

discriminatory conduct capable of excluding competitors or exploiting market power.

15. Autonomous Pricing Systems

Imagine thousands of firms using autonomous pricing systems.

Each algorithm observes:

• competitors;
• demand;
• inventories;
• market conditions;
• consumer responses.

The systems continuously modify prices.

The resulting concern is not necessarily an explicit cartel.

Instead:

Independent algorithms → repeated observation → rapid reaction → stable high-price equilibrium

This challenges traditional concepts of:

• agreement;
• intent;
• communication;
• foreseeability.

RealPage demonstrates why competition authorities are increasingly interested in algorithmic pricing systems.

16. Competition Governance of Foundation Models

Foundation models may become critical economic infrastructure.

Competition concerns can arise from:

Vertical integration

A company may control:

chips → cloud → foundation model → AI assistant → marketplace

Exclusive arrangements

A model provider may enter exclusive arrangements with:

• cloud providers;
• hardware manufacturers;
• distributors;
• application developers.

Data access

Competitors may lack equivalent training data.

Compute access

Training large models requires substantial computational resources.

Distribution

An AI model integrated into a dominant operating system may enjoy substantial advantages.

Therefore, competition governance should monitor vertical foreclosure across the entire AI stack.

17. Competition Remedies for Post-Human Markets

Traditional remedies may not always be sufficient.

A. Structural remedies

Possible remedies include:

• divestiture;
• separation of business units;
• prohibition of certain acquisitions.

B. Behavioural remedies

These can include:

• non-discrimination;
• transparency;
• data-access obligations;
• interoperability;
• restrictions on self-preferencing.

C. Technical remedies

These may involve:

• APIs;
• open standards;
• portability;
• interoperability protocols;
• algorithmic auditing.

D. Data remedies

Authorities could require:

• data portability;
• separation of datasets;
• restrictions on use of competitor data;
• access to certain datasets where legally justified.

E. Algorithmic remedies

Possible measures include:

• independent audits;
• monitoring;
• documentation;
• logging;
• restrictions on certain inputs;
• testing for discriminatory exclusion.

18. Human Agency and Consumer Choice

Post-human competition law must preserve the ability of humans to exercise meaningful economic choice.

Suppose an AI assistant automatically selects:

• a bank;
• an insurance company;
• a retailer;
• a hospital;
• a transport provider.

The consumer may technically retain choice, but practically the algorithm controls the choice architecture.

Competition governance therefore needs to consider:

Who controls the decision-making intermediary?

The relevant competitive actor may no longer be merely the seller.

It may be the AI system that decides which sellers consumers ever see.

19. Competition Neutrality Between Humans and Machines

A post-human economy may contain:

• human firms;
• autonomous firms;
• AI agents;
• robotic enterprises;
• decentralized organizations.

Competition law should remain technologically neutral.

The relevant question should generally be:

Does the conduct distort competitive conditions?

rather than:

Was the decision made by a human or a machine?

This prevents firms from escaping liability merely because a problematic decision was automated.

20. Emerging Doctrine: From Firm-Centric to System-Centric Competition Law

Traditional competition law often focuses on:

Firm A vs Firm B

Post-human competition requires consideration of:

Human + AI + Data + Platform + Cloud + Infrastructure + Algorithm + Ecosystem

The unit of competitive analysis may therefore increasingly become the economic system or ecosystem.

This does not mean abandoning firm-level competition law. Instead, it means adding an ecosystem perspective where conventional market boundaries are inadequate.

21. Major Competition Issues

22. Regulatory Challenges

1. Attribution

Who is legally responsible for an autonomous system's decision?

Possible answers include:

• the firm deploying the system;
• the system developer;
• the data provider;
• the platform operator;
• or multiple actors.

2. Explainability

Competition authorities need sufficient information to understand:

• why a price changed;
• why a competitor was demoted;
• why a supplier was excluded;
• why an algorithm recommended a particular product.

3. Speed

AI markets can change extremely quickly.

A conventional investigation may take years, while an AI market can experience substantial competitive transformation in months.

4. Technical complexity

Competition authorities increasingly require:

• data scientists;
• economists;
• AI engineers;
• cybersecurity specialists;
• forensic analysts.

5. International enforcement

AI markets are inherently cross-border.

A single model can operate simultaneously in:

• India;
• the EU;
• United States;
• China;
• Singapore;
• Middle East; and
• Africa.

Competition governance therefore requires international cooperation.

23. Emerging Principle of Algorithmic Competitive Neutrality

A useful conceptual principle for post-human competition law is:

Algorithmic competitive neutrality requires dominant digital systems to avoid using their computational control, data advantages or algorithmic position to unjustifiably distort competitive opportunities for dependent competitors.

It incorporates:

1. non-discrimination;
2. interoperability;
3. transparency;
4. fair access;
5. restrictions on self-preferencing;
6. appropriate data governance;
7. auditability.

Google Shopping provides a foundation for thinking about algorithmic self-preferencing, while the newer DMA framework explicitly addresses gatekeeper obligations concerning self-preferencing, steering and interoperability.

24. Six-Case Comparative Framework

25. Future Direction of Competition Governance

A mature post-human competition regime is likely to develop around several principles:

Principle 1 — Technological neutrality

The law should apply irrespective of whether decisions are human or algorithmic.

Principle 2 — Ecosystem accountability

Competition authorities should examine interconnected digital ecosystems rather than isolated products where appropriate.

Principle 3 — Data contestability

Dominant control over competitively important data should receive competition scrutiny.

Principle 4 — Algorithmic independence

Competitors should not use common systems in ways that unlawfully eliminate independent competitive decision-making.

Principle 5 — Interoperability

Where justified by market conditions, interoperability can prevent technological lock-in.

Principle 6 — Competitive neutrality

Dominant platforms should not unfairly favour their own downstream services.

Principle 7 — Innovation protection

Merger control should consider potential and innovation competition.

Principle 8 — Human choice

Consumers should retain meaningful ability to choose rather than being invisibly directed by dominant algorithmic intermediaries.

26. Conclusion

Competition law in post-human economic systems represents the transition from traditional firm-centred antitrust toward governance of complex technological ecosystems.

The central problem is no longer simply whether one company has market power. It is increasingly whether a company controls the combination of:

data + algorithms + AI models + compute + infrastructure + interfaces + distribution + autonomous decision-making.

The Google Shopping, Google Android, Amazon Marketplace, Meta Marketplace, Epic Games v Apple and RealPage matters collectively illustrate important dimensions of this transformation. They demonstrate competition concerns involving algorithmic ranking, ecosystem leverage, platform control, data advantages, distribution restrictions and algorithmic pricing.

The future challenge is therefore to construct a competition framework in which automation does not become a means of escaping antitrust responsibility, while innovation is not unnecessarily restrained by regulation.

In a genuinely post-human economic system, the fundamental competition-law question may ultimately become: