Competition Law and Intelligent Information Ecosystems and Dominance

1. Introduction

An intelligent information ecosystem is a digital environment in which data, algorithms, artificial intelligence, cloud infrastructure, search systems, recommendation engines, platforms, applications, advertising networks, identity systems and users interact continuously.

Examples include:

• search and information platforms;
• social-media ecosystems;
• e-commerce and marketplace platforms;
• AI and generative-AI ecosystems;
• digital advertising systems;
• app stores and operating systems;
• cloud and data-infrastructure platforms;
• financial-information and fintech ecosystems;
• health-information platforms; and
• interconnected IoT and smart-device ecosystems.

Competition law becomes particularly important where one undertaking controls several interconnected layers of such an ecosystem. Dominance may arise not merely from market share, but from the combination of data, network effects, switching costs, interoperability control, algorithms, infrastructure and access to users.

Modern competition authorities increasingly examine ecosystems rather than viewing every digital service as an isolated market. The European Commission's current market-definition materials expressly recognise digital ecosystems consisting of interconnected products and services.

2. Meaning of Intelligent Information Ecosystems

An intelligent information ecosystem can be understood as a system having five principal components:

A. Data layer

The platform collects:

• search histories;
• purchasing information;
• location data;
• behavioural data;
• social connections;
• transaction data;
• device information;
• advertising interactions; and
• information generated by third-party applications.

The accumulation of data can create a competitive advantage because more data may improve targeting, recommendation and prediction.

B. Algorithmic layer

Algorithms determine:

• search rankings;
• recommendations;
• advertising allocation;
• product visibility;
• pricing;
• content distribution;
• matching of buyers and sellers; and
• access to platform users.

C. Infrastructure layer

This may include:

• cloud infrastructure;
• operating systems;
• app stores;
• APIs;
• payment systems;
• identity systems;
• authentication;
• data centres; and
• network infrastructure.

D. User layer

Users generate additional data and network effects.

A simplified cycle is:

More users → more data → better algorithms → better service → more users → stronger market power.

E. Complementor layer

Third-party businesses depend on the ecosystem for:

• access to customers;
• app distribution;
• advertising;
• payment processing;
• cloud hosting;
• APIs;
• data access; and
• visibility.

This creates the possibility that the ecosystem operator may become both platform operator and competitor.

3. Why Intelligent Information Ecosystems Create Competition Concerns

3.1 Network effects

Digital ecosystems frequently exhibit direct or indirect network effects.

For example:

More sellers → more products → more consumers → more sellers.

Once a platform becomes sufficiently large, competitors may find it difficult to reproduce the same network.

3.2 Data advantages

Data can operate as an important competitive input.

A dominant platform may possess:

• larger datasets;
• higher-quality behavioural information;
• real-time information;
• cross-service data;
• historical datasets; and
• superior ability to combine datasets.

The competitive concern becomes stronger where rivals cannot reasonably obtain equivalent data.

3.3 Feedback loops

AI systems can create powerful feedback mechanisms:

Users → data → training → improved AI → better recommendations → more users → additional data.

Such feedback loops can reinforce an existing dominant position.

3.4 Switching costs

Users may hesitate to move because changing platforms requires:

• transferring data;
• rebuilding social connections;
• changing applications;
• learning another interface;
• losing accumulated reputation;
• abandoning purchase histories; or
• changing connected devices.

Switching costs therefore may make dominance more durable.

4. Relevant Market Definition

Traditional competition law begins with the definition of the relevant product and geographic market.

In intelligent information ecosystems, this becomes complicated because a single ecosystem can contain several interconnected markets.

For example:

Search → advertising → browser → operating system → mobile devices → app distribution → payments.

The authority must determine whether these constitute:

1. separate relevant markets;
2. interconnected markets; or
3. components of a broader ecosystem.

Chinese courts have expressly recognised the difficulty of market definition for comprehensive Internet platforms. In Shenzhen Weiyuanma Software Development Co. v Tencent, the Supreme People's Court-related case materials emphasised that the relevant product market must consider the particular service to which the allegedly abusive conduct is directed and distinguish basic services from value-added services.

5. Dominance in Intelligent Information Ecosystems

Dominance may be assessed through traditional factors supplemented by digital-market characteristics.

Traditional factors

• market share;
• financial strength;
• barriers to entry;
• countervailing buyer power;
• technological advantages;
• control of essential inputs.

Digital factors

• network effects;
• data advantages;
• ecosystem integration;
• interoperability;
• switching costs;
• default settings;
• algorithmic control;
• access to APIs;
• control over app stores;
• control over user identity;
• ecosystem lock-in; and
• ability to leverage power into neighbouring markets.

Thus:

Market share ≠ the entire dominance analysis.

A firm with a relatively moderate share may nevertheless possess strategic control over a bottleneck such as data, an operating system, a marketplace or an API.

6. Major Competition Concerns

A. Self-preferencing

An ecosystem operator may favour its own services over those of competitors.

Examples include:

• placing its own products first;
• preferential search rankings;
• preferred recommendation placement;
• superior access to APIs; or
• preferential advertising treatment.

This was central to the Google Shopping litigation.

B. Data leveraging

A dominant platform may use data collected in one market to strengthen its position in another.

Example:

Messaging data → advertising data → targeted advertising → stronger advertising position.

This theory was important in the Meta proceedings.

C. Data tying

The platform may condition access to one service upon acceptance of extensive data collection or combination.

This raises both competition and data-protection concerns.

D. Refusal of interoperability

A dominant ecosystem may restrict access to:

• APIs;
• technical interfaces;
• operating systems;
• data portability;
• interoperability tools; or
• essential technical standards.

Such conduct can increase barriers to entry.

E. Bundling and tying

An ecosystem may combine:

• operating system + search;
• search + browser;
• marketplace + payments;
• app store + payment system;
• hardware + software.

Bundling becomes particularly significant where the dominant product gives the firm a gateway into another market.

F. Exclusive dealing

A platform may require sellers, developers or advertisers to avoid competing ecosystems.

This can reinforce network effects and foreclose rivals.

G. Algorithmic discrimination

Algorithms may selectively:

• downgrade competitors;
• raise competitors' costs;
• favour affiliated businesses;
• allocate visibility;
• manipulate recommendations; or
• disadvantage multi-homing.

7. Important Case Laws

1. Google Search (Shopping) — European Commission / General Court

Case: Google and Alphabet v European Commission, Google Shopping litigation.

Google was found to have favoured its own comparison-shopping service in search-result placement over competing comparison-shopping services.

The case is significant because it demonstrates how control over information discovery and ranking can become a source of market power.

The General Court upheld the Commission's finding of an abuse, although aspects of the Commission's reasoning were modified.

Principle

A dominant information intermediary cannot necessarily use control over an important gateway to systematically disadvantage competing services.

2. Google Android — European Commission

Case: Google Android, Case AT.40099 / subsequent General Court litigation.

The case concerned Google's practices involving Android, including restrictions connected with search and browser distribution and contractual arrangements with device manufacturers.

The importance for intelligent information ecosystems is the interaction between:

operating system + search + browser + app distribution + mobile devices.

The case illustrates ecosystem leveraging, where power in one technological layer can influence competition in another.

The EU's current digital-market framework continues to treat Google's Search, Android, Chrome, Maps, Play and other services as interconnected core platform services.

3. Meta Platforms v Bundeskartellamt

Case: Meta Platforms Inc. and Others v Bundeskartellamt, C-252/21, judgment of 4 July 2023.

This is one of the most important cases concerning data-driven dominance.

The German competition authority challenged Meta's combination of Facebook user data with so-called off-Facebook data, including information obtained from third-party websites and services.

The Court of Justice considered the relationship between competition law and GDPR requirements.

Principle

Competition authorities may need to consider data-protection law when assessing whether a dominant platform's data-processing practices constitute abusive conduct.

The case demonstrates that data governance can become a competition-law issue when data exploitation is connected to market power.

4. Meta/Facebook Marketplace — European Commission

Case: Commission Case AT.40684.

The Commission's investigation concerned Meta's conduct relating to Facebook Marketplace.

A significant theory concerned the interaction between:

Facebook → user base → Marketplace → advertising data → competing marketplace services.

The Commission's later market-definition materials identify Facebook Marketplace as an ecosystem-related digital competition case.

The case demonstrates the possibility of leveraging power from one digital service into another.

5. Amazon Marketplace / Amazon Buy Box — European Commission and UK CMA

Amazon's ecosystem illustrates a different information problem.

Amazon possesses extensive information concerning third-party sellers, including information about:

• sales;
• inventory;
• performance; and
• marketplace activity.

The UK CMA investigated whether Amazon's use of seller data could give Amazon an advantage in competing with those sellers. Commitments included restrictions concerning non-public seller data and changes concerning Buy Box allocation.

Principle

A platform may face competition concerns when it simultaneously:

1. operates the marketplace;
2. collects commercially sensitive information from participants; and
3. competes against those same participants.

This creates a platform-as-regulator versus platform-as-competitor problem.

6. Qihoo 360 v Tencent — China

Case: Beijing Qihoo Technology Co. Ltd. v Tencent Holdings / Tencent Technology.

The litigation arose from conflict between Qihoo 360 and Tencent concerning Tencent's QQ ecosystem and software compatibility.

The Chinese court considered:

• market definition;
• free digital services;
• network effects;
• dominance;
• bundled software;
• compatibility; and
• restrictions affecting competing software.

The case is particularly important because conventional SSNIP-style market analysis can become difficult where the consumer price is zero. The court's analysis highlighted the need to examine the actual characteristics and substitution possibilities of Internet services.

Principle

Zero monetary price does not mean absence of a relevant competition market.

Competition analysis must examine functionality, substitution, users, technology and network characteristics.

7. Alibaba — China

Case: Alibaba Group Holding Ltd., SAMR decision, 2021.

SAMR found Alibaba responsible for an abuse involving its "choose one from two" exclusivity practice concerning merchants.

The case is significant for information ecosystems because Alibaba's platform possessed substantial:

• merchant information;
• consumer information;
• transaction information;
• platform traffic; and
• network effects.

The decision was accompanied by a substantial monetary penalty, equivalent to 4% of Alibaba's domestic turnover for 2019.

Principle

Platform dominance can be reinforced through contractual restrictions that prevent merchants from effectively participating in competing ecosystems.

8. Meituan — China

The Chinese enforcement against Meituan concerned the platform's use of exclusivity arrangements in relation to merchants.

The case is relevant to intelligent information ecosystems because platforms can use:

• transaction data;
• merchant information;
• consumer traffic;
• algorithmic recommendations; and
• platform dependence

to reinforce their market position.

The broader Chinese enforcement approach has increasingly examined the relationship between platform economics, data and exclusionary conduct.

8. Cross-Market Leveraging

One of the most important theories in intelligent ecosystems is:

Dominance in Market A can create competitive advantages in Market B.

For example:

Messaging dominance
↓
Large user dataset
↓
Advertising intelligence
↓
Superior targeting
↓
Advertising-market advantage.

The Indian competition-law proceedings involving Meta illustrate this theory. The 2025 CCI decision discussed how data generated through WhatsApp could potentially reinforce Meta's position in online display advertising and affect market access for competitors.

This illustrates why ecosystem competition cannot always be analysed by examining each service independently.

9. The Data Advantage as a Competitive Moat

Data can create several forms of competitive advantage.

1. Scale advantage

More users generate more data.

2. Scope advantage

Data from multiple services can be combined.

3. Speed advantage

Real-time data may allow faster optimisation.

4. Quality advantage

A platform may possess richer behavioural information.

5. Learning advantage

Data improves machine-learning models.

Thus:

Data → AI improvement → better service → more users → more data.

This produces a self-reinforcing competitive loop.

10. Artificial Intelligence and Ecosystem Dominance

AI intensifies these concerns.

An AI ecosystem may contain:

• foundation models;
• training data;
• cloud computing;
• GPUs;
• model marketplaces;
• application programming interfaces;
• application stores;
• AI assistants;
• search;
• advertising; and
• user data.

Competition concerns may therefore arise when one company controls several layers.

For example:

Cloud infrastructure
↓
AI model access
↓
Developer APIs
↓
Applications
↓
Users
↓
Behavioural data

Control of several layers may create entry barriers for competing AI providers.

11. Ecosystem Lock-In

Lock-in occurs when users or businesses become dependent on the ecosystem.

Important mechanisms include:

• proprietary data formats;
• lack of portability;
• interoperability restrictions;
• loyalty programmes;
• accumulated reputation;
• subscriptions;
• connected devices;
• application compatibility;
• contractual restrictions; and
• network effects.

Competition law therefore increasingly examines whether a platform's conduct makes multi-homing or switching unnecessarily difficult.

12. Essential-Facility-Like Information Resources

In certain circumstances, an information resource may become strategically indispensable.

Potential examples include:

• technical APIs;
• interoperability information;
• certain platform data;
• authentication systems;
• app-distribution access;
• infrastructure interfaces.

However, not every valuable dataset constitutes an essential facility.

Authorities normally need to consider:

1. whether the resource is genuinely indispensable;
2. whether duplication is practically possible;
3. whether access refusal excludes competition;
4. whether access is technically feasible;
5. whether legitimate business justifications exist; and
6. whether access would impair incentives to innovate.

13. Competition Law and Data Protection

The Meta v Bundeskartellamt litigation demonstrates an important convergence.

Competition law asks:

Does the conduct distort competition?

Data protection law asks:

Is personal-data processing lawful and properly justified?

The two questions are legally distinct but may interact.

This is particularly important where:

dominance + compulsory data collection + cross-service data combination

occur together.

14. Ex-Ante Regulation

Traditional competition law is predominantly ex post:

Conduct → investigation → infringement finding → remedy.

Digital ecosystem regulation increasingly adds ex ante obligations.

The EU Digital Markets Act is the clearest example. The EU currently identifies Alphabet, Amazon, Apple, Booking, ByteDance, Meta and Microsoft as designated gatekeepers, covering numerous core platform services.

The policy rationale is that certain ecosystem structures may be difficult to correct through traditional abuse-of-dominance proceedings alone.

15. Possible Competition Remedies

Competition authorities may consider:

Structural remedies

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

Behavioural remedies

• non-discrimination;
• prohibition of self-preferencing;
• restrictions on exclusivity;
• transparent ranking;
• fair access.

Data remedies

• data portability;
• interoperability;
• data-sharing obligations;
• restrictions on combining datasets.

Technical remedies

• API access;
• interoperability;
• switching tools;
• open standards.

Governance remedies

• independent monitoring;
• algorithmic auditing;
• compliance reporting;
• transparent platform rules.

16. Key Legal Tests

A competition-law investigation into an intelligent information ecosystem can follow this framework:

Step 1 — Identify the ecosystem

↓

Step 2 — Identify the relevant services and markets

↓

Step 3 — Assess market power

↓

Step 4 — Identify data and infrastructure advantages

↓

Step 5 — Examine network effects and switching costs

↓

Step 6 — Identify the allegedly abusive conduct

↓

Step 7 — Examine cross-market leveraging

↓

Step 8 — Assess foreclosure or exploitative effects

↓

Step 9 — Consider efficiencies and legitimate justifications

↓

Step 10 — Select proportionate remedies

17. Distinctive Features of Intelligent Information Ecosystem Dominance

18. Key Principles Emerging from the Case Law

The cases collectively demonstrate several important propositions:

1. Data can be a source of market power.
2. A zero-price service can still constitute a relevant competition market.
3. Control over information ranking can produce exclusionary effects.
4. Dominance in one digital market may be leveraged into another.
5. Network effects can make digital dominance self-reinforcing.
6. Platform access can become strategically important for downstream competitors.
7. A platform competing with its own business users creates special information asymmetries.
8. Data protection and competition law can interact in assessing dominant-platform conduct.
9. Interoperability and portability can materially affect contestability.
10. Ecosystem power may require analysis extending beyond a single product market.

Recent scholarship likewise identifies theories such as blocking entry paths and defensive foreclosure as particularly relevant to ecosystem competition.

19. Conclusion

Competition law and intelligent information ecosystems represent an evolution from traditional market-centred antitrust analysis toward analysis of data, algorithms, infrastructure, network effects and interconnected digital services.

The central competition problem is not simply that a company becomes large. The critical question is whether control over information, data, infrastructure and ecosystem rules enables the undertaking to:

• exclude competitors;
• leverage dominance into adjacent markets;
• disadvantage dependent businesses;
• restrict interoperability;
• exploit informational advantages;
• prevent effective multi-homing; or
• make market entry and expansion substantially more difficult.

The most significant authorities and cases—including Google Shopping, Google Android, Meta v Bundeskartellamt, Meta/Facebook Marketplace, Amazon Marketplace, Qihoo v Tencent, Alibaba and Meituan—show different manifestations of this problem across jurisdictions. Together they demonstrate why modern competition law increasingly examines the architecture of digital ecosystems rather than merely the market share of an individual product.