Competition Law and Algorithmic Market Design Embedded in Operating Systems

1. Meaning

Algorithmic market design embedded in operating systems refers to the use of software rules, algorithms, default settings, technical interfaces, permissions, ranking systems, and automated decision-making mechanisms built directly into an operating system to influence how users, developers, applications, advertisers, and other businesses interact in the market.

Examples include:

default search engines;

default browsers;

app-store ranking;

application permissions;

notification controls;

payment systems;

advertising identifiers;

APIs;

interoperability rules;

security restrictions;

AI assistants;

recommendation systems;

application-installation controls;

background-process restrictions; and

automated allocation of device resources.

The competition-law question is not simply whether the operating system uses algorithms. The central question is:

Does the operating system's technical and algorithmic design legitimately improve the product, or is it being used by a firm with market power to restrict competition in neighbouring markets?

2. Why Operating Systems Are Important Competition Infrastructure

An operating system is more than software running a device.

It can function as a market gateway between:

Users ↔ Applications ↔ Developers ↔ Advertisers ↔ Payment providers ↔ Search engines ↔ Cloud services

Consequently, whoever controls the operating system may influence access to several downstream markets.

For example:

Operating system → default browser → search engine → advertising market.

A technical decision made at the OS level can therefore have consequences beyond the OS market itself.

3. Algorithmic Market Design

Traditional market design concerns the rules under which market participants interact.

An operating system can effectively perform this function through software.

For example, its algorithms may determine:

which application is displayed first;

which application is pre-installed;

which browser opens links;

which search engine receives queries;

which payment system is available;

which APIs are accessible;

which applications can operate in the background;

which notifications users receive;

which applications can access device functions.

Thus:

Code can function as a set of market rules.

4. Main Competition-Law Risks

A. Self-Preferencing

An operating-system owner may compete with third-party applications while controlling the technical environment in which those applications operate.

Possible examples:

preferential ranking;

privileged API access;

superior system integration;

pre-installation;

default status;

privileged notifications.

The concern is greatest where the OS owner has substantial market power.

5. Default Settings

Defaults are particularly powerful because many users never change them.

Examples:

default browser;

default search engine;

default maps application;

default payment service;

default assistant.

A default can generate:

automatic exposure → increased usage → more data → better service → stronger market position.

Therefore, an apparently minor technical choice can have substantial competitive consequences.

6. Pre-Installation

Pre-installation can provide a significant distribution advantage.

Suppose an operating system automatically installs:

OS owner's search + browser + assistant + cloud storage.

A rival must then persuade the user to:

discover its product;

download it;

install it;

change the default; and

continue using it.

This creates an asymmetric distribution burden.

7. Algorithmic Ranking

Operating systems may contain recommendation or ranking systems.

For example, an OS may recommend:

applications;

games;

subscriptions;

cloud services;

payment services;

security tools.

If the platform also owns competing products, the ranking system may become a potential channel of self-preferencing.

Relevant questions include:

What factors determine ranking?

Are proprietary services treated differently?

Are competing applications disadvantaged?

Can developers understand the relevant rules?

Does the algorithm change after rival entry?

8. API Access

Application programming interfaces allow third-party applications to interact with operating-system functions.

An OS provider may control access to:

location;

payments;

notifications;

Bluetooth;

cameras;

secure storage;

authentication;

voice assistants.

If the OS provider offers competitors restricted access while giving its own products superior access, competition concerns may arise.

However:

Not every API restriction is anticompetitive.

Security, privacy and technical-integrity reasons can justify restrictions.

9. Interoperability

An operating system can determine whether rival products can interact effectively with it.

Examples include:

messaging;

file transfer;

wearable devices;

payment systems;

cloud services;

smart-home products.

Reduced interoperability may increase ecosystem lock-in.

10. Security as a Competition Issue

Operating-system companies frequently justify restrictions on the basis of:

cybersecurity;

malware prevention;

privacy;

system stability.

These can be legitimate objectives.

Competition authorities therefore need to ask:

Is the restriction genuinely necessary for security, or is security being used as a justification for protecting a commercial product?

This is highly fact-specific.

11. Algorithmic Resource Allocation

An operating system can automatically decide:

CPU allocation;

battery priority;

network access;

background execution;

notification priority.

If the OS owner's application consistently receives technical advantages unavailable to competitors, this may potentially affect competition.

The authority would need evidence establishing:

the relevant market;

the OS provider's market power;

differential treatment;

competitive significance; and

lack of sufficient legitimate justification.

12. AI Assistants Inside Operating Systems

AI assistants create a new form of operating-system control.

An integrated assistant may become the user's gateway to:

search;

shopping;

travel;

applications;

financial services;

entertainment;

information.

Instead of users opening different applications, the assistant could determine which service receives the request.

For example:

User: “Find me a hotel.”

The AI assistant decides:

Which search provider → which booking service → which payment provider

This transforms the assistant into a potential algorithmic intermediary.

13. Operating-System Ecosystems and Network Effects

An OS can create a feedback loop:

More users

↓

More developers

↓

More applications

↓

Better ecosystem

↓

More users

↓

More developers

This can make entry difficult.

If the OS owner also competes with developers, the owner may have incentives to use ecosystem control strategically.

14. Tying

Operating systems can also facilitate tying.

For example:

OS → search service

or

OS → browser

or

OS → payment system

or

OS → cloud service

The legal assessment depends on the applicable jurisdiction and the evidence concerning market power, tying conditions, foreclosure and legitimate justifications.

15. Bundling

An ecosystem provider can bundle:

operating system;

cloud storage;

security;

browser;

search;

AI assistant;

productivity software.

Bundling can produce genuine efficiencies.

But it can also make it harder for independent providers to compete for individual components.

16. Algorithmic Switching Barriers

An operating system can make switching more difficult by controlling:

data migration;

account systems;

application compatibility;

device pairing;

authentication;

cloud synchronisation.

A competitor may therefore face a technical switching barrier, even when consumers are legally free to switch.

17. Data Advantages

Operating systems can collect information concerning:

application usage;

device behaviour;

location;

interaction patterns;

purchases;

device configuration.

When an OS provider also operates competing services, data may provide a competitive advantage.

Competition authorities may therefore examine:

what data is collected;

who can access it;

whether rivals receive comparable information;

whether data is combined across services;

whether data creates a durable advantage.

18. Case Law

The following cases are especially relevant. Some concern operating systems directly; others establish principles applicable to algorithmic OS design.

Case 1: United States v Microsoft

United States v Microsoft Corp., 253 F.3d 34 (D.C. Cir. 2001)

Facts

Microsoft possessed substantial power in PC operating systems and was accused of using that position to restrict competition from web browsers.

Relevance

The case is foundational for understanding how an operating-system platform can influence an adjacent market.

The court considered:

operating-system control;

technical integration;

distribution;

contractual restrictions;

browser competition.

Principle

A dominant platform can potentially use control over a technological environment to disadvantage an adjacent competitor.

Modern application

The same reasoning can be relevant when an OS owner controls:

APIs;

defaults;

application distribution;

technical permissions;

AI assistants.

19. Case 2: Google Android

Google and Alphabet v Commission, Case T-604/18, General Court, 2022

This is one of the most directly relevant modern cases.

Facts

The case concerned Google's Android ecosystem and contractual arrangements involving mobile-device manufacturers and application distribution.

Ecosystem structure

Android connected:

Operating system → devices → applications → search → users

Competition relevance

The case demonstrates that restrictions connected with an operating system can influence competition in adjacent markets.

Algorithmic market-design relevance

Modern OS competition may involve:

defaults;

pre-installation;

application access;

interoperability;

technical integration.

The Android case therefore provides an important framework for examining ecosystem-based competition concerns.

20. Case 3: Google Shopping

Google and Alphabet v Commission (Google Shopping), Case C-48/22 P, CJEU, 2024

Facts

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

Relevance

Search algorithms can determine whether competing services receive meaningful user traffic.

Principle

The competitive significance of a digital platform can extend to the way its systems:

rank;

display;

position; and

distribute visibility.

OS application

An operating system can similarly control application visibility through:

recommendation systems;

search;

app stores;

default interfaces.

Thus, algorithmic ranking can function as a competitive gatekeeper.

21. Case 4: Epic Games v Apple

Epic Games, Inc. v Apple Inc., 67 F.4th 946 (9th Cir. 2023)

Facts

The dispute involved Apple's App Store ecosystem, including application distribution and payment rules.

Relevance

Apple's operating-system ecosystem connected:

iOS;

App Store;

developers;

consumers;

payment systems.

Competition significance

The case demonstrates how control over the operating environment and application distribution can create important competitive issues.

Algorithmic-design relevance

The modern issue extends beyond contractual rules to:

application discovery;

ranking;

payment interfaces;

technical permissions;

interoperability.

22. Case 5: Intel v Commission

Intel Corp. v Commission, Case C-413/14 P, CJEU, 2017

Principle

The case concerned rebates provided by Intel to computer manufacturers and a retailer.

The CJEU held that where the undertaking provides evidence that the conduct is not capable of restricting competition, the Commission may need to consider the relevant economic circumstances in assessing the foreclosure capability of the conduct.

Relevance

An OS provider may use commercial incentives together with technical advantages.

Therefore investigators should not examine:

technical restriction + contractual incentive

in complete isolation when their competitive effects may interact.

23. Case 6: Bronner

Oscar Bronner GmbH & Co. KG v Mediaprint, Case C-7/97, CJEU, 1998

Principle

The case established demanding conditions concerning when refusal to provide access to an infrastructure controlled by a dominant undertaking may constitute abuse.

Relevance to operating systems

Modern questions can involve:

API access;

interoperability;

technical interfaces;

operating-system functionality.

Important limitation

Bronner does not establish that dominant OS providers must generally provide every technical facility to competitors.

The exceptional conditions for compulsory access remain important.

24. Case 7: Magill

RTE and ITP v Commission, Joined Cases C-241/91 P and C-242/91 P, CJEU, 1995

Principle

Magill concerns exceptional circumstances surrounding refusal to license information protected by intellectual-property rights.

Relevance

Operating systems contain valuable proprietary:

APIs;

technical information;

interfaces;

software components.

The case provides historical context for analysing exceptional circumstances where control over proprietary information may become relevant to competition.

Limitation

It should not be read as creating a general obligation to disclose proprietary software technology.

25. Case 8: Microsoft v Commission

Microsoft Corp. v Commission, Case T-201/04, General Court, 2007

Facts

The case involved Microsoft's Windows operating system and interoperability information concerning work-group server products.

Importance

This is especially relevant to operating-system competition.

The General Court upheld the Commission's approach concerning Microsoft's refusal to provide interoperability information and the relationship between Windows' dominance and neighbouring server markets.

Principle

Control over a dominant technological platform can create competition concerns where access or interoperability restrictions prevent effective competition in an adjacent market and the applicable legal conditions are satisfied.

Modern relevance

The same analytical framework can be relevant to:

APIs;

cloud interfaces;

AI assistants;

device interoperability;

cross-platform functionality.

26. Case 9: Meta Platforms v Bundeskartellamt

Meta Platforms Inc. and Others v Bundeskartellamt, Case C-252/21, CJEU, 2023

Relevance

The case demonstrates how digital platform competition can intersect with:

data;

user behaviour;

platform services;

data-processing practices.

For operating systems, similar questions can arise when an OS provider combines data from:

applications;

devices;

accounts;

advertising;

cloud services.

The case therefore supports a broader understanding of data-related ecosystem power.

27. Algorithmic Market Design vs Traditional Product Design

There is an important difference.

Traditional product design

The company decides:

“What features should our product contain?”

Algorithmic market design

The company effectively decides:

“What rules should govern how users and competing businesses interact with our platform?”

The second can have greater competition significance because the software itself establishes the conditions of market participation.

28. Potential Competition Problems

These are potential concerns, not automatic infringements.

29. Legitimate Reasons for OS Restrictions

An effective competition analysis must consider legitimate explanations.

Operating-system restrictions may be justified by:

Security

Preventing malware.

Privacy

Protecting user data.

Reliability

Preventing system crashes.

Battery management

Reducing excessive resource consumption.

Quality control

Ensuring compatibility.

Fraud prevention

Protecting payment systems.

Consumer protection

Preventing misleading applications.

Therefore, a competition authority should distinguish:

genuine technical necessity

from

commercially motivated exclusion.

30. Proportionality Analysis

A useful framework is:

Question 1

What competitive restriction exists?

Question 2

What legitimate objective is claimed?

Question 3

Is the restriction technically capable of achieving that objective?

Question 4

Is there a less restrictive alternative?

Question 5

Does the restriction disproportionately disadvantage competing products?

This approach is particularly relevant to:

security controls;

API restrictions;

app-store rules;

interoperability.

31. AI Assistants as New Operating-System Gatekeepers

The future competition issue may be particularly significant.

Previously:

User → App → Service

Increasingly:

User → AI assistant → Service

If the AI assistant is embedded into the operating system, the OS provider may control the intermediate layer.

The assistant could determine:

which application receives a request;

which search engine is used;

which retailer is shown;

which payment method is offered;

which information is prioritised.

This creates a new form of algorithmic intermediation.

32. Competition Extinction Risk

If the same company controls:

Operating system + AI assistant + app store + search + payments + cloud

then independent competitors may face several simultaneous disadvantages.

For example:

Rival app

↓ restricted API

↓ less functionality

↓ lower ranking

↓ fewer users

↓ less data

↓ weaker AI recommendations

↓ fewer developers

↓ weaker ecosystem

This demonstrates how multiple small restrictions could potentially reinforce one another.

33. Regulatory Investigation Framework

Competition authorities investigating an OS should examine five layers.

Layer 1 — Architecture

What does the operating system control?

Layer 2 — Algorithm

How does the relevant algorithm make decisions?

Layer 3 — Access

How do third-party competitors interact with the system?

Layer 4 — Economics

What effect does the system have on:

price;

quality;

innovation;

entry;

switching?

Layer 5 — Legal test

Does the conduct satisfy the relevant competition-law standard?

34. Evidence Authorities May Need

An algorithmic investigation may require:

source code;

API documentation;

system logs;

algorithm versions;

A/B test results;

ranking data;

developer communications;

internal strategy documents;

technical specifications;

consumer data;

app-distribution data.

This is why modern competition authorities increasingly require computer scientists, data scientists and digital-forensics experts alongside lawyers and economists.

35. Remedies

Where unlawful conduct is established, possible remedies could include:

Choice screens

Allowing users to choose competing services.

Interoperability

Requiring technically meaningful compatibility.

Non-discrimination

Preventing the OS from giving its own competing product unjustified advantages.

API access

Providing appropriate access to competing services.

Data portability

Helping consumers move relevant data.

Default reforms

Giving users genuine choice.

Monitoring

Independent monitoring of algorithmic changes.

Structural remedies

In exceptional circumstances, structural separation may be considered under the relevant legal framework.

36. Important Limitation

Competition law should not treat every technical integration as suspicious.

Modern operating systems are necessarily integrated.

Integration can create:

efficiency;

security;

innovation;

lower costs;

better user experience.

The relevant distinction is therefore:

Efficient technological integration vs. exclusionary use of technological control.

37. Comparative Legal Principles

EU approach

Particular importance may be placed on:

dominance;

foreclosure;

self-preferencing;

tying;

interoperability;

platform regulation;

effects on competitive structure.

U.S. approach

Analysis commonly focuses on:

monopolization;

exclusionary conduct;

competitive effects;

consumer welfare;

technological integration;

legitimate business justifications.

The precise legal test varies according to the jurisdiction and conduct.

38. Short Hypothetical

Suppose OS-A controls 80% of a mobile operating-system market.

It also owns Assistant-A, which competes with independent AI assistants.

OS-A's algorithm:

automatically activates Assistant-A;

gives Assistant-A privileged system access;

restricts rival assistants' background operation;

places Assistant-A above rivals;

prevents rivals from accessing certain APIs.

The authority should not simply conclude:

“OS-A is dominant, therefore the conduct is unlawful.”

Instead it should examine:

relevant market;

dominance;

precise restrictions;

technical necessity;

foreclosure;

consumer effects;

innovation effects;

efficiencies;

less restrictive alternatives.

That is the appropriate competition-law framework.

39. Conclusion

Algorithmic market design embedded in operating systems represents an important evolution of digital competition law.

Operating systems can function as technical markets and gateways, determining:

who gets access;

what users see;

which applications are available;

how applications interact;

which services become defaults;

how data flows;

how payments occur; and

how AI assistants interact with competing services.

The most relevant legal precedents include Microsoft, Google Android, Google Shopping, Epic Games v Apple, Intel, Bronner, Magill, Microsoft interoperability, and Meta Platforms.

The central principle is:

An operating system's technical architecture can become competition-relevant when a firm with substantial market power uses that architecture to influence access, distribution, interoperability, defaults, ranking or other competitive conditions in neighbouring markets.

At the same time, technical integration, security restrictions, proprietary design and ecosystem development are not automatically anticompetitive. The legal assessment must focus on the specific conduct, market power, competitive effects, legitimate justifications and applicable jurisdictional standards.

Exam Keywords

Operating system – algorithmic market design – digital platform – ecosystem – defaults – pre-installation – self-preferencing – API access – interoperability – tying – bundling – foreclosure – leveraging – app store – AI assistant – ranking algorithm – network effects – switching costs – data advantage – platform gatekeeper – Microsoft – Google Android – Google Shopping – Epic Games – Intel – Bronner – Magill – interoperability – technical integration – algorithmic governance – digital competition.