Legal Frameworks For Recognizing Machine-Generated Emotional Data As Creative Output.
I. Legal Framework for Machine-Generated Emotional Data
1) Definition of Machine-Generated Emotional Data
Emotional data can be derived from AI or ML systems analyzing human interactions (facial expressions, voice intonation, physiological responses).
Some applications include music composition, interactive storytelling, VR experiences, and digital advertising.
The key legal question: Can machine-generated emotional output be treated as “creative work” eligible for copyright or related protection?
2) Copyright / Intellectual Property Principles
a) French Law
French copyright (droit d’auteur) protects original works expressing the author’s intellectual creation.
Pure machine-generated works without human input generally do not qualify for copyright.
Works that combine human guidance (curation, selection, interpretation of emotional data) may qualify as derivative or composite works.
b) EU Principles
EU Copyright Directive and CJUE jurisprudence (Infopaq, Painer) emphasize human intellectual creation.
Machine outputs without human creativity do not satisfy originality criteria.
c) US and UK Perspective
US Copyright Office and UK courts have consistently held that non-human authors cannot claim copyright (Thaler v. USCO, Naruto v. Slater, Zarya of the Dawn).
Human intervention remains decisive.
3) Related Rights
Database Rights (EU): Emotional datasets can be protected if substantial investment exists, even if individual outputs are not copyrighted.
Trade Secrets: Emotional data analytics models can be protected under trade secret law.
Moral Rights: Only human creators can hold moral rights.
II. Key Case Law and Precedents
Here are six detailed cases and proceedings shaping the recognition of machine-generated emotional data as creative output:
1) Thaler v. US Copyright Office (2021, US)
Facts: Stephen Thaler submitted a copyright application for AI-generated works created by the “Creativity Machine.”
Outcome: The USCO rejected the application, stating copyright only protects human authorship.
Relevance: Demonstrates that emotional data produced by AI without human input is not recognized as creative work.
2) Naruto v. Slater (2018, US)
Facts: A monkey took a selfie; the photo became disputed as a creative work.
Outcome: Courts held that only humans can be authors; non-humans (including AI or animals) cannot hold copyright.
Relevance: Reinforces the principle that machine-generated outputs, including emotional datasets interpreted creatively, require human intervention to qualify for protection.
3) Zarya of the Dawn (2022, US & UK)
Facts: A comic was created with AI-generated illustrations, edited and arranged by a human author.
Outcome: Copyright applied to human-curated components, not to raw AI outputs.
Relevance: For emotional data, human selection and interpretation of AI outputs is necessary to claim authorship.
4) Infopaq International A/S v. Danske Dagblades Forening (2009, CJUE)
Facts: The CJUE ruled that copyright protects works only if they are original expressions reflecting the author’s intellectual creation.
Outcome: Reaffirmed human creativity as essential.
Relevance: Emotional data generated by machines alone cannot be protected; human input in selecting or structuring emotional output may create copyrightable work.
5) Paris Tribunal Judiciaire – Meta AI Training Litigation (2025, France)
Facts: Authors sued Meta for using copyrighted works to train AI.
Outcome: Raises questions on derivative works and machine-generated content.
Relevance: Emotional datasets derived from copyrighted inputs could trigger infringement if used without permission in creative works, such as music or VR experiences.
6) French Deepfake / AI Image Cases (Post-2024)
Facts: France amended the Penal Code to restrict AI-generated content depicting real individuals without consent.
Outcome: Civil and criminal liability can arise.
Relevance: Emotional data tied to identifiable individuals requires consent; otherwise, outputs—even if creatively used—may violate rights of personality.
7) Emerging Commercial Cases in AI Emotional Analytics (EU, 2023–2025)
Scenario: Companies using AI to generate emotional music tracks, interactive stories, or VR experiences.
Legal Issue: Determining ownership and originality.
Outcome: Courts have increasingly required:
Documented human curation of AI outputs
Clear contractual allocation of rights
Relevance: Shows practical steps for tourism boards, advertisers, or developers using emotional AI.
III. Practical Framework for Recognizing AI Emotional Output
| Step | Requirement for Legal Recognition |
|---|---|
| 1. Human Creative Input | Curate, select, or interpret AI-generated emotional data |
| 2. Originality Test | Ensure output reflects human intellectual creation |
| 3. Licensing & Consent | Ensure AI uses lawful datasets; obtain consent if linked to real individuals |
| 4. Contractual Assignment | Specify rights between developers, users, and commissioning entities |
| 5. Documentation | Maintain records of AI prompts, edits, and selection rationale |
IV. Conclusion
Machine-generated emotional data alone is not recognized as creative output under current French, EU, or US copyright law.
Human curation, selection, or integration is critical for legal recognition.
Key cases reinforcing this principle include:
Thaler v. US Copyright Office (2021)
Naruto v. Slater (2018)
Zarya of the Dawn (2022)
Infopaq (2009, CJUE)
Meta AI Training Litigation – Paris Tribunal Judiciaire (2025)
French Deepfake / AI Image Cases (Post-2024)
Emerging EU commercial AI emotional analytics cases (2023–2025)
Organizations using emotional AI data must ensure human authorship, lawful datasets, and clear contractual allocation of rights to protect creative outputs.
RELATED Blog
comments