Ambient Computing and Passive Data Extraction Dominance

Introduction

Ambient computing refers to computing systems embedded into everyday environments so that technology operates continuously, often with limited deliberate interaction by the user. Examples include smart speakers, connected vehicles, wearable devices, smart-home systems, location services, voice assistants, IoT sensors, connected televisions, health devices, and AI-enabled environmental sensors.

Passive data extraction occurs when these systems collect, infer, combine, or monetize information generated through ordinary use—such as location, voice interactions, device identifiers, behavioural patterns, contacts, browsing activity, biometric information, purchasing behaviour, and sensor data—without requiring a separate, conscious data-submission event for each item.

From a competition-law perspective, the central issue is not merely privacy. Persistent data collection can become a source of market power, entry barriers, ecosystem dependence, exclusionary conduct, and competitive advantage. A dominant platform may use accumulated data to improve algorithms, target advertising, personalize services, discriminate between users or suppliers, strengthen network effects, and make switching more difficult.

The legal analysis therefore connects competition law, data governance, consumer protection, interoperability, privacy, and digital-platform regulation.

1. Meaning of Ambient Computing

Ambient computing generally has five characteristics:

1. Continuous operation – devices remain active in the background.
2. Low-friction interaction – users do not have to initiate every data transaction.
3. Sensor dependence – microphones, cameras, GPS, accelerometers and other sensors generate information.
4. Cross-service integration – data from multiple products can be combined.
5. Algorithmic processing – collected data can be transformed into behavioural predictions or commercial insights.

For competition law, the important distinction is between:

Data voluntarily supplied for a particular transaction
and
Data continuously generated as a by-product of participation in an ecosystem.

The latter can create significant competitive advantages where competitors cannot obtain comparable datasets.

2. What Is Passive Data Extraction?

Passive data extraction may include:

• location tracking;
• voice recordings and voice-command metadata;
• device telemetry;
• browsing and search histories;
• purchasing behaviour;
• biometric or physiological information;
• movement patterns;
• smart-home sensor data;
• connected-car information;
• app usage;
• contacts and social graphs;
• inferred interests;
• household composition;
• behavioural predictions; and
• cross-device identifiers.

The competition concern becomes particularly significant when a firm can aggregate data across otherwise separate markets.

For example:

Smart speaker → voice data → user preferences → advertising profile → targeted advertising → improved advertising revenues → acquisition of more devices → more data.

This can generate a data-driven feedback loop.

3. Competition-Law Theory of Harm

A. Data as a competitive input

Data can function as an important input where:

• its quantity matters;
• quality matters;
• freshness matters;
• competitors cannot easily replicate it;
• data has network or learning effects; and
• access to it materially improves products or services.

A dominant firm controlling a unique dataset may therefore possess an important competitive advantage.

However, possession of large amounts of data does not automatically establish dominance. Authorities generally need to examine market definition, substitutability, barriers to entry, data replicability and the actual competitive effects.

4. Data-Driven Network Effects

Ambient devices can create a self-reinforcing cycle:

More users

↓

More passive data

↓

Better algorithms / personalization

↓

Better service

↓

More users

↓

Still more data

This is sometimes described as a data-network effect or data feedback loop.

A competitor entering later may face a structural disadvantage because it does not possess the historical dataset necessary to train or optimize comparable systems.

5. Ecosystem Lock-In

Ambient computing frequently involves ecosystems rather than individual products.

For example:

Smartphone → smartwatch → smart speaker → connected television → automobile → cloud account

may all operate through one identity and data architecture.

A consumer may therefore face substantial switching costs.

The competition question is whether the firm uses this ecosystem to:

• restrict interoperability;
• prevent data portability;
• degrade compatibility;
• tie products together;
• restrict rival applications;
• make competing devices technically inferior; or
• make accumulated data unavailable to rivals.

6. Passive Data and Privacy as a Competition Parameter

Privacy can sometimes function as a quality dimension of competition.

If two services are nominally free, consumers may compare them according to:

• amount of data collected;
• transparency;
• ability to control collection;
• retention practices;
• third-party sharing;
• security;
• ability to delete or transfer data.

Consequently, a reduction in privacy protection can potentially constitute a deterioration in product quality.

Competition authorities therefore increasingly examine the relationship between data practices and competitive parameters.

7. Relevant Market Definition

Several markets may need examination simultaneously.

Device market

Examples:

• smart speakers;
• wearables;
• connected vehicles;
• smart televisions.

Platform market

Examples:

• voice-assistant platforms;
• mobile operating systems;
• smart-home ecosystems.

Data market

Potentially involving:

• advertising data;
• location information;
• behavioural datasets;
• identity data.

Advertising market

Ambient data can strengthen:

• targeted advertising;
• attribution;
• behavioural advertising;
• retail media.

The same conduct may therefore produce effects across multiple interconnected markets.

8. Abuse of Dominance

Potential theories include:

8.1 Refusal of data access

A dominant platform may refuse competitors access to commercially important data.

8.2 Discriminatory access

A platform may provide data or APIs to its own services on better terms than to competitors.

8.3 Self-preferencing

The platform may use information collected from third-party suppliers to favour its own downstream service.

8.4 Tying and bundling

Access to an important dataset may be conditioned on adoption of another product.

8.5 Exclusivity

Manufacturers may be prevented from supporting competing ecosystems.

8.6 Degrading interoperability

A dominant firm may technically impair compatibility with rival devices.

8.7 Exploitative data collection

In exceptional circumstances, excessive data extraction may become relevant to competition analysis where it reflects the exercise of market power.

9. Data Advantage and Barriers to Entry

A new competitor may need:

• millions of observations;
• historical behavioural information;
• real-time sensor information;
• labelled datasets;
• training data;
• user interaction data.

The incumbent may have accumulated these datasets over many years.

This creates a potential data-entry barrier.

The important legal question is whether the dataset is:

1. unique;
2. difficult to reproduce;
3. necessary or highly valuable;
4. continuously refreshed; and
5. capable of conferring durable competitive advantages.

10. Six Important Case Laws

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

The Google Shopping litigation concerned Google's treatment of competing comparison-shopping services in its search results.

Although the case did not specifically concern ambient computing, it is highly relevant to the broader theory of leveraging dominance in one digital ecosystem into adjacent services.

The case demonstrates how a platform controlling an important digital gateway may affect downstream competitors through the manner in which its infrastructure operates.

Relevance: Ambient platforms could similarly use control over a device, operating system, voice assistant or data gateway to favour affiliated services.

2. Google Android — European Commission

The Android proceedings concerned Google's contractual arrangements involving Android devices, including practices concerning search, browsers and application distribution.

The case illustrates how a powerful digital ecosystem can employ contractual and technical arrangements across interconnected products.

Relevance to ambient computing: A dominant smart-device ecosystem could potentially use similar arrangements to reinforce its position in connected devices, voice assistants, applications or data services.

3. Google Search (AdSense) — European Commission

The AdSense case concerned contractual restrictions relating to Google's advertising intermediation services.

It is relevant because it illustrates the competition-law importance of control over an intermediary and restrictions affecting access to competing services.

Ambient-computing relevance: If an ambient-computing platform controls both the collection of behavioural information and advertising monetisation, restrictions imposed on rival advertising or data services could raise analogous concerns.

4. Bundeskartellamt — Facebook / Meta Data Combination Case

The German competition authority examined Facebook's combination of user data originating from Facebook with data from other services and external sources.

The case is especially important because it demonstrated the intersection between data practices and competition law.

The authority's theory connected Facebook's market position with its ability to impose data-related conditions on users.

Ambient-computing relevance: A company operating connected devices could potentially combine information generated across multiple services and devices, increasing the competitive significance of its data advantage.

5. Bundeskartellamt — Google / Alphabet Data and Conditions Proceedings

The German proceedings involving Google examined Google's ability to combine and process data across different Google services and the implications of those practices for competition.

The proceedings are particularly relevant to the concept of cross-service data combination.

Ambient-computing relevance: Smart speakers, search, maps, advertising, smartphones and connected-device services can generate complementary datasets. Combining them can strengthen the platform's informational advantage.

6. FTC v. Facebook

The U.S. litigation involving Facebook addressed alleged anticompetitive conduct concerning Facebook's position in personal social networking and its treatment of competing services.

Although not an ambient-computing case, it is relevant to the broader issue of platform power, network effects and strategies affecting actual or potential competitors.

Ambient-computing relevance: A connected-device ecosystem may similarly create network effects and switching costs that reinforce an incumbent's position.

11. Additional Relevant Authorities

7. Google Shopping — ECJ

The later European judicial proceedings concerning Google Shopping are significant for understanding how conduct involving a dominant digital platform can be assessed where the platform gives preferential treatment to its own downstream service.

This is relevant to self-preferencing using information generated within a platform ecosystem.

8. Apple App Store / App Store Competition Proceedings

Competition investigations and litigation concerning Apple's App Store model have examined issues including payment restrictions, platform access and the relationship between Apple and competing digital services.

The broader principle is relevant to ambient ecosystems because a device manufacturer may simultaneously control:

• hardware;
• operating systems;
• application distribution;
• payment infrastructure;
• user identity; and
• data flows.

That vertical integration can make interoperability particularly important.

12. Essential-Facilities Dimension

A difficult issue is whether an ambient platform's dataset could constitute an essential facility.

Traditional essential-facilities analysis generally requires careful examination of factors such as:

1. indispensability;
2. absence of realistic alternatives;
3. duplication feasibility;
4. exclusionary effects; and
5. justification for refusal.

Not every valuable database qualifies.

For ambient computing, the strongest case would potentially arise where a platform controls a uniquely valuable dataset or technical interface that competitors cannot reasonably reproduce.

13. Interoperability as a Competition Remedy

Competition authorities may consider remedies such as:

• API access;
• interoperability obligations;
• data portability;
• technical access;
• non-discrimination obligations;
• separation of data pools;
• restrictions on cross-use of data;
• transparency obligations; and
• restrictions on self-preferencing.

The objective is not necessarily to require unrestricted disclosure of all data.

Instead, remedies may attempt to prevent strategic use of informational advantages to exclude competitors.

14. Data Portability

Data portability can reduce switching costs.

Suppose a user has accumulated:

• voice preferences;
• smart-home configurations;
• fitness history;
• vehicle settings;
• purchase histories;
• location information.

If none of this can be transferred to a competing ecosystem, the consumer may be effectively locked in.

Portability can therefore operate as a competition-enhancing mechanism.

15. Algorithmic Advantages

Passive data becomes particularly valuable when used to train algorithms.

Consider:

100 million devices

→ continuous behavioural observations

→ machine-learning dataset

→ improved prediction

→ improved personalization

→ higher engagement

→ greater advertising value

→ increased revenue

→ additional device acquisition.

The resulting advantage can become difficult for competitors to replicate.

16. Discrimination Against Rival Devices

A dominant ambient platform might theoretically:

• reduce functionality on rival hardware;
• delay API access;
• restrict voice-assistant integration;
• limit sensor access;
• degrade synchronization;
• impose discriminatory certification requirements.

Such conduct could raise concerns under abuse-of-dominance or exclusionary-conduct rules where the necessary dominance and competitive effects are established.

17. Privacy Degradation as Quality Degradation

One of the most important emerging theories is:

Competition can occur over privacy quality as well as monetary price.

If users cannot meaningfully negotiate data collection because one platform dominates the relevant market, privacy conditions may deteriorate without consumers having realistic alternatives.

Competition authorities may therefore examine:

• data minimisation;
• consent mechanisms;
• collection frequency;
• cross-use of data;
• retention periods;
• third-party sharing; and
• user control.

18. Consumer Exploitation Versus Exclusion

It is important to distinguish two theories.

Exploitative theory

The dominant company uses market power to impose excessively intrusive data conditions.

Exclusionary theory

The dominant company uses accumulated data to disadvantage competitors.

The second theory is generally easier to connect directly to conventional competition analysis, while the first raises more complex questions concerning the role of privacy and consumer protection.

19. Merger-Control Implications

Ambient computing also raises data-driven merger concerns.

Suppose a dominant connected-device company acquires:

• a voice-recognition provider;
• a smart-home platform;
• a health-data company;
• an advertising technology company; or
• a connected-car data provider.

Authorities may examine whether the transaction combines datasets in a manner that:

• increases entry barriers;
• eliminates an emerging competitor;
• facilitates cross-market leveraging;
• increases advertising concentration; or
• reduces privacy competition.

Traditional turnover thresholds may be inadequate where an innovative data-rich company generates relatively little revenue.

20. Killer-Acquisition Concerns

A large technology platform could potentially acquire a smaller company before its technology or dataset becomes commercially significant.

Competition analysis may therefore examine:

• nascent competition;
• innovation pipelines;
• alternative technologies;
• unique datasets;
• potential competition; and
• future ecosystem expansion.

The central issue is whether the transaction removes an important potential competitive constraint.

21. Tacit Coordination Risks

Ambient systems can generate highly granular real-time information.

For example:

Real-time demand data → algorithmic pricing → competitor observation → automated response → synchronized prices

This can create competition concerns if independent firms use algorithms or information-sharing mechanisms that facilitate coordination.

However, parallel algorithmic behaviour alone does not necessarily establish an unlawful agreement. Authorities must distinguish legitimate independent adaptation from conduct amounting to prohibited coordination.

22. Relevant Competition-Law Questions

A regulator investigating an ambient-computing platform could ask:

Market power

• What is the relevant market?
• How many viable alternatives exist?
• Can consumers realistically multi-home?

Data

• What data is collected?
• How frequently?
• Is it unique?
• Can competitors reproduce it?

Interoperability

• Can rival devices connect?
• Are APIs available?
• Are technical standards discriminatory?

Switching

• Can users transfer historical data?
• Are devices compatible with competing services?

Vertical integration

• Does the platform compete with businesses dependent on it?

Exclusion

• Does the platform use data obtained from rivals to compete against them?

Privacy

• Is reduced privacy a competitive-quality issue?

Remedies

• Would portability, interoperability or access obligations restore competitive conditions?

23. Compliance Framework for Businesses

Companies operating ambient-computing systems should establish:

1. Data inventory
2. Purpose limitation
3. Access controls
4. Competitor-neutral API policies
5. Interoperability review
6. Non-discrimination procedures
7. Data portability mechanisms
8. Internal competition-law training
9. Algorithmic auditing
10. Merger-risk assessment

Particular attention should be given to situations where the company both controls the data infrastructure and competes with firms dependent upon that infrastructure.

24. Emerging Competition-Law Risks

25. Key Legal Principles Emerging From the Cases

The case law collectively supports several propositions:

Principle 1

Data can be a source of competitive advantage without being a conventional market commodity.

Principle 2

Control over a digital gateway can permit leverage into adjacent markets.

Principle 3

Cross-service data combination can strengthen ecosystem power.

Principle 4

Interoperability and access can become important competition parameters.

Principle 5

Privacy-related conditions may sometimes form part of competitive-quality analysis.

Principle 6

Network effects and switching costs can reinforce incumbent positions.

Principle 7

A competition analysis must distinguish legitimate data-driven innovation from exclusionary use of data.

Conclusion

Ambient computing transforms data collection from an occasional transaction into a continuous feature of economic participation. That transformation has major competition-law implications.

The central concern is not simply that a company possesses large quantities of information. The critical question is whether persistent data extraction, aggregation and algorithmic exploitation create or reinforce market power in a way that disadvantages competitors, increases switching costs, restricts interoperability, or enables leveraging across related markets.

The combination of ambient devices + continuous data + AI + ecosystem integration + network effects can therefore produce a powerful feedback mechanism:

More devices → more passive data → better algorithms → stronger services → greater ecosystem adoption → more data.