Attention Manipulation Liability in European Law

1. Meaning of Attention Manipulation Liability

Attention Manipulation Liability concerns legal responsibility for using digital systems, interface design, algorithms, behavioural profiling, advertising, recommender systems, or other techniques to capture, retain, redirect, or exploit a person's attention in a way that causes legally recognizable harm.

It is closely connected with the attention economy but is narrower.

The distinction is:

• Attention economy = the broader economic model of competing for and monetizing attention.
• Attention manipulation = techniques used to influence where, when, and how a person directs attention.
• Attention manipulation liability = legal responsibility arising when those techniques violate an applicable legal duty and cause legally recognized harm.

There is no autonomous European cause of action called "attention manipulation liability."

A claim must normally be constructed through one or more existing legal regimes, including:

• GDPR;
• consumer protection;
• Digital Services Act;
• competition law;
• employment law;
• discrimination law;
• contract;
• tort/delict;
• product or service liability;
• privacy;
• fundamental rights.

2. Examples of Attention Manipulation

Attention manipulation may involve:

Interface manipulation

• infinite scrolling;
• autoplay;
• hidden cancellation buttons;
• repeated consent requests;
• confusing privacy settings;
• preselected options.

Behavioural manipulation

• personalized notifications;
• variable rewards;
• streaks;
• gamification;
• algorithmically selected content;
• emotionally provocative recommendations.

Advertising manipulation

• behavioural advertising;
• personalized advertising;
• hidden commercial content;
• influencer marketing;
• targeted advertising based on inferred characteristics.

Algorithmic manipulation

Algorithms may predict:

• what a user will click;
• what content will retain the user;
• what advertisement will produce a response;
• when a user is most likely to return.

The legal issue becomes particularly serious where the system is designed to exploit:

• vulnerability;
• addiction-like behaviour;
• children's immaturity;
• psychological characteristics;
• lack of information.

3. Core Legal Principle

The mere fact that a business attempts to persuade someone does not normally create liability.

Advertising and persuasion are legitimate commercial activities.

Liability becomes more plausible where there is:

Manipulative conduct + breach of a legal duty + legally protected interest + causation + legally recognized harm.

For example:

Personalized advertising

by itself does not necessarily create liability.

But:

unlawful profiling → manipulative targeting → significant financial decision → legally recognized loss

may create a potential claim, depending on the applicable law.

4. Main Sources of European Liability

4.1 GDPR

The GDPR is central when attention manipulation relies upon personal data.

Relevant concepts include:

• lawful processing;
• fairness;
• transparency;
• purpose limitation;
• data minimization;
• profiling;
• consent;
• automated decision-making;
• data-subject rights;
• accountability.

5. Fairness Under the GDPR

Article 5(1)(a) GDPR requires processing to be:

• lawful;
• fair;
• transparent.

This is particularly relevant to attention manipulation.

A platform may technically disclose that it uses data while still raising questions concerning whether the processing is genuinely fair and sufficiently understandable.

For example:

"We use your activity to personalize your experience"

may be insufficient if the actual system involves extensive behavioural profiling and individualized persuasion that the user could not reasonably understand.

6. Transparency

Attention-manipulation systems can be extremely complex.

The user may not know:

• why particular content appears;
• why a particular advertisement appears;
• why a notification was sent;
• how the system predicts behaviour;
• which data was used;
• whether the content is commercially sponsored.

European data-protection law therefore places significant emphasis on meaningful information.

7. Case Law: Orange România

Orange România SA, C-61/19

The CJEU considered whether consent had genuinely been given under the GDPR.

The Court emphasized that consent must be:

• freely given;
• specific;
• informed;
• unambiguous.

Attention-manipulation relevance

A platform cannot manufacture consent through interface design.

For example, consent may be problematic where:

• options are preselected;
• refusing is unnecessarily difficult;
• information is hidden;
• the user is pressured into accepting.

The case is not directly about attention manipulation but is an important authority for interface-based consent manipulation.

8. Case Law: Planet49

Planet49 GmbH, C-673/17

The CJEU considered consent in relation to cookies.

The Court rejected the idea that a pre-ticked box could automatically establish valid consent.

Principle

Silence, inactivity, or a preselected option does not necessarily demonstrate genuine consent.

Attention-manipulation relevance

Interface architecture cannot be used to manufacture legally meaningful consent.

This is one of the strongest analogical authorities for dark-pattern litigation.

9. Case Law: Wirtschaftsakademie

Wirtschaftsakademie Schleswig-Holstein, C-210/16

The case concerned a Facebook fan page and responsibility for personal-data processing.

The CJEU recognized joint responsibility in circumstances where the operator participated in determining purposes and means of processing.

Attention-manipulation relevance

Attention manipulation frequently involves several actors:

website → platform → advertising network → analytics provider → advertiser.

A business cannot necessarily avoid responsibility simply because another company operates the underlying technical system.

10. Case Law: Fashion ID

Fashion ID GmbH & Co. KG v Verbraucherzentrale NRW, C-40/17

The case concerned a social-media plugin embedded on a website.

The CJEU examined the responsibility of the website operator for personal-data processing associated with the plugin.

Principle

Participation in a data-processing ecosystem can create legal responsibility even without control over every aspect of the technology.

Attention relevance

A publisher that embeds:

• tracking pixels;
• social plugins;
• advertising technology;
• recommendation tools;

may have legal responsibilities concerning the resulting processing.

11. Case Law: Google Spain

Google Spain SL and Google Inc. v AEPD and Costeja González, C-131/12

The CJEU recognized important data-protection rights concerning search-engine processing.

Principle

Search engines exercise significant influence over the accessibility and presentation of information concerning individuals.

Attention-manipulation relevance

Search rankings determine:

• what users see first;
• what receives attention;
• what is ignored;
• how information is discovered.

Google Spain therefore demonstrates that algorithmic information intermediaries can have legally significant effects on individuals.

It is an analogical authority rather than a direct attention-manipulation case.

12. Case Law: SCHUFA

SCHUFA Holding (Scoring), C-634/21

The CJEU addressed automated scoring under Article 22 GDPR.

The Court treated automated scoring as potentially significant where it effectively determines important decisions about an individual.

Principle

The legal consequences of an algorithm cannot be determined merely by looking at the technical label assigned to the process.

Attention-manipulation relevance

The case provides a broader algorithmic principle:

Where automated systems substantially influence individual outcomes, legal safeguards may become more important.

Again, SCHUFA is not a direct attention-manipulation case.

13. Case Law: Delfi AS v Estonia

Delfi AS v Estonia, ECtHR Grand Chamber, 2015

Delfi was an online news platform that hosted user comments.

The ECtHR considered whether holding the platform liable for unlawful comments violated freedom of expression.

The Court accepted liability in the specific circumstances.

Attention relevance

The case demonstrates that digital intermediaries can bear legal responsibility for the consequences of online systems.

But the case also illustrates an important limitation:

Platform liability must be balanced against freedom of expression.

Therefore, regulation of attention systems cannot ignore Article 10 ECHR.

14. Case Law: MTE and Index.hu v Hungary

MTE and Index.hu v Hungary, ECtHR, 2016

The ECtHR examined liability for online comments and emphasized the importance of context and proportionality.

Principle

Digital-platform liability cannot be assessed mechanically.

Courts must consider:

• nature of the speech;
• context;
• platform role;
• consequences;
• measures taken;
• freedom of expression.

Attention-manipulation relevance

Recommendation systems and engagement algorithms influence what speech receives visibility.

Any regulation must therefore balance:

protection from manipulation/harm

against

freedom to receive and impart information.

15. Consumer Law

Attention manipulation can also constitute an unfair commercial practice.

European consumer law is concerned with practices that materially distort the economic behaviour of consumers.

Examples may include:

• fake countdown timers;
• misleading scarcity;
• hidden costs;
• disguised advertisements;
• subscription traps;
• difficult cancellation;
• false social proof;
• emotionally manipulative interfaces.

The key legal question is:

Did the commercial practice materially distort or materially impair the consumer's ability to make an informed transactional decision?

16. Dark Patterns

Dark patterns are particularly important.

A dark pattern may:

• conceal information;
• make one option substantially easier than another;
• repeatedly pressure the user;
• exploit cognitive biases;
• make cancellation unnecessarily difficult;
• manipulate privacy choices.

Not every inconvenient interface is unlawful.

The legal analysis depends upon:

1. what the interface does;
2. the relevant legal duty;
3. the user's vulnerability;
4. whether the practice is commercial;
5. whether it materially distorts behaviour.

17. Digital Services Act

The DSA is particularly relevant to attention manipulation.

It addresses platform design and user protection, including restrictions on dark patterns.

It also establishes obligations concerning:

• recommender systems;
• advertising transparency;
• protection of minors;
• systemic risks;
• platform accountability.

This represents an important shift.

Traditional consumer law often asks:

"Was a particular commercial practice unfair?"

The DSA can additionally ask:

"Does the design and operation of the platform create systemic risks or undermine users' ability to make free and informed choices?"

18. Recommender Systems

Recommendation algorithms can shape attention by determining:

• which videos appear;
• which posts appear;
• which products appear;
• which news stories appear;
• which accounts are recommended.

The algorithm may optimize for:

• engagement;
• watch time;
• clicks;
• advertising revenue;
• retention.

This can create legal concerns where optimization systematically promotes harmful or manipulative outcomes.

19. Children's Attention

Children are particularly important.

Children may have greater difficulty understanding:

• profiling;
• targeted advertising;
• algorithmic recommendations;
• commercial persuasion;
• long-term privacy consequences.

Attention manipulation aimed at children can therefore raise concerns under:

• GDPR;
• DSA;
• consumer law;
• Charter Article 24;
• national child-protection law.

20. Profiling Children

Suppose a platform learns that a 13-year-old user:

• repeatedly watches appearance-related videos;
• responds to weight-loss advertisements;
• engages with emotionally distressing content.

If the platform uses this information to optimize engagement or advertising, serious legal questions may arise concerning:

• profiling;
• children's vulnerability;
• fairness;
• targeted advertising;
• platform safety.

The precise legality depends upon the applicable legal framework and factual circumstances.

21. Fundamental Rights

Attention manipulation can affect several fundamental rights.

Article 7 EU Charter

Privacy and communications.

Article 8 EU Charter

Personal-data protection.

Article 11 EU Charter

Freedom of expression and information.

Article 24 EU Charter

Rights of the child.

Article 38 EU Charter

Consumer protection.

Article 47 EU Charter

Effective judicial protection.

Under the ECHR, relevant rights can include:

• Article 8 — private life;
• Article 10 — expression and information;
• Article 14 — non-discrimination.

22. Autonomy

A major theoretical issue is personal autonomy.

The law generally accepts persuasion.

For example:

"This product may be useful to you."

is ordinary advertising.

But increasingly personalized systems may attempt to predict:

"This particular person is most susceptible to this particular message at this particular time."

The closer the system moves toward exploiting vulnerability rather than informing choice, the stronger the legal concerns may become.

23. Manipulation vs Persuasion

This distinction is fundamental.

The boundary is fact-sensitive.

24. Competition Law

Attention manipulation can also produce competition concerns.

A dominant platform may control:

• search visibility;
• app distribution;
• advertising;
• recommendation systems;
• social feeds.

It could potentially use that control to:

• favour its own services;
• disadvantage competitors;
• exploit business users;
• restrict rival access;
• combine data across services.

Competition law therefore focuses less on individual psychological harm and more on market power and competitive structure.

25. Google Shopping

Google and Alphabet v Commission, C-48/22 P

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

Attention relevance

Search ranking determines the distribution of:

• visibility;
• clicks;
• traffic;
• commercial opportunities.

Thus, algorithmic control over attention can have competition-law significance.

The case should nevertheless be described as a competition-law analogy, not a direct attention-manipulation authority.

26. Discrimination Through Attention Allocation

An algorithm might determine that certain advertisements should be shown to certain people.

This can produce discriminatory outcomes.

For example:

• housing advertisements;
• employment advertisements;
• credit offers;
• educational opportunities.

A platform may unintentionally or deliberately distribute opportunities differently according to:

• sex;
• age;
• nationality;
• disability;
• ethnic origin;
• other protected characteristics or proxies.

Relevant European equality authorities include:

CHEZ Razpredelenie Bulgaria, C-83/14

The CJEU examined indirect discrimination and discriminatory effects.

D.H. and Others v Czech Republic

The ECtHR addressed indirect discrimination in education.

These are analogical authorities, not attention-manipulation decisions.

27. Employment Context

Attention manipulation can also arise in workplaces.

Employers may use:

• productivity notifications;
• engagement dashboards;
• algorithmic scheduling;
• digital monitoring;
• attention metrics;
• employee-screen monitoring.

The legal issues may involve:

• privacy;
• employment law;
• data protection;
• proportionality;
• collective rights.

Bărbulescu v Romania

The ECtHR emphasized safeguards when employers monitor employees' communications.

López Ribalda and Others v Spain

The ECtHR examined covert workplace surveillance.

These cases are relevant by analogy to workplace attention monitoring.

28. Tort and Civil Liability

A claimant might attempt to use national tort law where attention manipulation produces actual harm.

Possible allegations include:

• negligence;
• breach of statutory duty;
• intentional interference;
• misrepresentation;
• nuisance;
• product/service defects.

However, the claimant generally must prove:

1. duty;
2. breach;
3. causation;
4. legally recognized damage.

This is much more difficult than simply showing:

"The platform kept me online longer than I intended."

29. Addiction-Like Harm

One of the most difficult emerging questions is whether excessive platform engagement can amount to compensable harm.

Potential claims might allege:

manipulative design → compulsive engagement → psychological/economic/social harm.

The major legal obstacles are:

• proving the duty;
• proving manipulation;
• proving causation;
• distinguishing ordinary use from legally actionable harm;
• establishing recoverable damage;
• proving that the platform's conduct was legally unreasonable.

There is currently no general European rule saying that a platform is automatically liable because its service is highly engaging.

30. Causation

Causation is particularly difficult.

Suppose:

Algorithm → increased screen time → sleep disruption → health problem.

The claimant may need to establish:

• the algorithm caused the increased engagement;
• the engagement caused the relevant behaviour;
• that behaviour caused the injury;
• the injury is legally attributable to the defendant.

Alternative causes can include:

• individual behaviour;
• other platforms;
• employment;
• family circumstances;
• pre-existing conditions;
• independent lifestyle factors.

Consequently, scientific and expert evidence may become important.

31. Economic Harm

Attention manipulation may also produce financial harm.

Examples:

• impulsive purchases;
• gambling-related spending;
• repeated subscription payments;
• in-app purchases;
• deceptive upselling;
• targeted high-pressure advertising.

A claimant may have a stronger case where there is an identifiable:

manipulative practice → transaction → financial loss

chain.

32. Gambling and High-Risk Commercial Environments

Attention manipulation becomes particularly sensitive where platforms facilitate:

• gambling;
• financial products;
• high-risk investments;
• credit;
• addictive commercial services.

In these areas, sector-specific regulation may impose stronger obligations.

The same interface technique that is relatively harmless in a music application may be far more problematic when used to induce repeated gambling expenditure.

33. Liability of Different Actors

Potential defendants include:

Platform operator

Responsible for:

• interface;
• recommendation algorithms;
• data processing;
• advertising systems.

Advertiser

Responsible for:

• commercial message;
• targeting;
• misleading claims.

Data broker

May be responsible for:

• unlawful profiling;
• unlawful data acquisition;
• disclosure.

App developer

May be responsible for:

• manipulative design;
• unlawful processing;
• contractual violations.

Publisher

May be responsible for:

• embedded tracking;
• commercial content;
• misleading presentation.

Employer

May be responsible for:

• workplace attention monitoring.

Liability depends upon the specific legal duty of each actor.

34. Joint Responsibility

Modern digital ecosystems often involve multiple participants.

For example:

User → social network → analytics company → advertising exchange → advertiser.

Cases such as Wirtschaftsakademie and Fashion ID demonstrate why responsibility may sometimes be shared.

However, joint controllership or participation in data processing does not automatically mean that every participant is liable for every form of harm.

The scope of responsibility must be determined separately.

35. Defences

Possible defences include:

1. No manipulation

The platform may argue that the design merely provides ordinary personalization.

2. No legal duty

The claimant has not identified a legally enforceable obligation.

3. Valid consent

The defendant may rely on valid consent where legally permissible.

4. No significant effect

The defendant may argue that the processing did not produce legally significant consequences.

5. Lack of causation

The alleged manipulation did not cause the claimed injury.

6. No compensable damage

The claimant suffered annoyance or dissatisfaction but no legally recoverable loss.

7. Freedom of expression

The platform may invoke Article 10 ECHR or Article 11 Charter where the claim affects dissemination of lawful content.

8. Legitimate commercial interest

The defendant may argue that personalization or advertising serves legitimate commercial purposes, subject to the applicable legal restrictions.

36. Remedies

Depending on the cause of action, remedies can include:

Data-protection remedies

• access;
• rectification;
• erasure;
• restriction;
• objection;
• cessation of unlawful processing;
• compensation where applicable.

Consumer remedies

• cancellation;
• restitution;
• injunction;
• compensation where available.

Regulatory remedies

• administrative fines;
• compliance orders;
• corrective measures;
• restrictions on processing or advertising.

Civil remedies

• damages;
• injunctions;
• declarations;
• restitution.

37. Six Core Case Laws

For a six-case academic answer, the strongest authorities are:

1. Planet49 — C-673/17

Consent and manipulative interface design.

2. Orange România — C-61/19

Free and informed consent.

3. Wirtschaftsakademie — C-210/16

Platform responsibility and joint controllership.

4. Fashion ID — C-40/17

Responsibility within digital data-processing ecosystems.

5. Google Spain — C-131/12

Digital intermediary power and personal-data rights.

6. SCHUFA — C-634/21

Consequential automated processing and algorithmic decision-making.

For a broader treatment, add:

1. Delfi AS v Estonia — platform liability and freedom of expression.
2. MTE and Index.hu v Hungary — proportionality and online intermediary liability.
3. Google Shopping / Google and Alphabet — algorithmic visibility and competition.
4. CHEZ — discrimination by apparently neutral systems.

38. Consolidated Case Table

39. A Practical Legal Test for Attention Manipulation Liability

A useful European legal analysis can proceed through ten questions:

1. What technique was used?

Was it:

• profiling;
• autoplay;
• notifications;
• targeted advertising;
• recommender systems;
• dark patterns;
• gamification?

2. What data was used?

Determine whether the system processes:

• personal data;
• behavioural data;
• sensitive data;
• inferred characteristics.

3. What was the purpose?

Was the purpose:

• advertising;
• engagement;
• retention;
• sales;
• political communication;
• personalization?

4. What legal basis exists?

For personal data, identify the applicable GDPR basis.

5. Was the user adequately informed?

Assess transparency.

6. Was choice genuinely free?

Apply the principles from Planet49 and Orange România.

7. Was the practice commercially manipulative?

Apply consumer-protection principles.

8. Did the system create discriminatory or harmful effects?

Consider equality and fundamental-rights law.

9. Did the conduct cause legally recognizable harm?

Establish:

• financial loss;
• privacy harm;
• other protected injury.

10. What remedy is legally available?

Separate:

• regulatory remedies;
• data-protection remedies;
• consumer remedies;
• civil damages.

40. Central Doctrinal Problem

The greatest challenge in attention-manipulation litigation is distinguishing:

ordinary persuasion

from

legally impermissible manipulation.

European law does not generally establish a right to be completely free from persuasive influence.

Instead, it protects particular interests:

• informational self-determination;
• privacy;
• personal-data protection;
• consumer autonomy;
• equality;
• freedom of expression;
• children's interests;
• property;
• effective judicial protection.

Consequently, a claimant normally needs to connect attention manipulation to one of these legally protected interests.

41. Regulatory Violation Does Not Automatically Equal Civil Liability

This is especially important.

Suppose a regulator concludes:

"A platform breached a DSA obligation."

That does not necessarily establish that every user can sue the platform for damages.

Likewise:

GDPR infringement ≠ automatic proof of financial loss.

A private damages action may still require:

• an actionable legal basis;
• infringement;
• causation;
• compensable damage.

The distinction between public regulatory liability and private civil liability should therefore always be maintained.

42. Conclusion

Attention Manipulation Liability in European law is an emerging, cross-disciplinary field rather than an autonomous cause of action. Liability generally arises by connecting manipulative attention practices to an existing legal duty under data-protection, consumer, digital-services, competition, equality, employment, tort, or fundamental-rights law.

The most important principles are:

1. Personalized attention systems may constitute profiling under the GDPR.
2. Consent cannot be manufactured through deceptive interface design.
3. Planet49 and Orange România are central authorities on genuine consent.
4. Wirtschaftsakademie and Fashion ID demonstrate that responsibility can extend across digital ecosystems.
5. Google Spain illustrates the legal significance of powerful digital intermediaries.
6. SCHUFA demonstrates the increasing importance of safeguards surrounding consequential automated processing.
7. Delfi and MTE show that platform regulation must be balanced against freedom of expression.
8. Competition law can address manipulation of digital visibility where market power is involved.
9. Children and vulnerable users may require enhanced protection.
10. Most importantly, unlawful manipulation does not automatically establish a damages claim; the claimant must identify a legal duty, breach, causation and legally recognizable harm.

The emerging European approach can therefore be summarized as:

The law does not prohibit platforms from attracting attention; it increasingly regulates the data, algorithms, interfaces and commercial techniques used to influence attention when those techniques undermine meaningful choice, privacy, consumer autonomy, equality, safety or other legally protected interests.

Available next action: Create a downloadable PDF file here in this chat containing the findings and recommendations above