Transparency
Audiences should not be misled about whether marketing depicts real people, experiences or endorsements. Disclosure should address material risks of misunderstanding rather than rely on vague statements about AI use.
Creativity & intellectual property IWIB4AI session Needs review
Marketing and Content in the Age of AI – Legal and Ethical Considerations
Using generative AI for brand content while respecting copyright, informing audiences and avoiding unfair marketing.
Draft – legal and case law references not yet verified
Summary
Generative AI can help brands create marketing copy and visual content, but its use raises questions about intellectual property, audience transparency and fairness. Responsible adoption requires checking rights in source material and outputs, assessing whether disclosures are needed, and reviewing content for misleading claims or harmful stereotypes.
The Problem
Brands must balance faster content production with uncertainty about rights, authenticity and the effects of AI-generated messaging on audiences. Neither an AI tool's availability nor its contractual terms alone establish that an output is lawful, protectable or fair to use.
Ethical Grounds
Fairness
AI-generated marketing can reproduce stereotypes or systematically exclude groups from brand representation. Review should consider who is depicted, how they are portrayed and whether messaging exploits particular vulnerabilities.
Autonomy
Synthetic testimonials, imagery or claims can undermine consumers' ability to make informed choices. Brands should preserve a clear distinction between creative presentation and factual evidence about products or experiences.
Accountability
Brands should assign responsibility for rights clearance, claim verification and publication approval. Using an external AI tool does not remove the need for internal review or a process to correct harmful content.
Legal Compliance
| Framework | Provision | Relevance |
|---|---|---|
| EU copyright law | Directive 2001/29/EC, Arts. 2–3 | Using protected works as inputs or publishing outputs that reproduce protected expression can engage reproduction and communication-to-the-public rights, requiring permission unless an applicable exception applies. |
| EU consumer law | Directive 2005/29/EC, Arts. 5–7 | Consumer-facing AI-generated marketing must not constitute an unfair commercial practice, including misleading actions or omissions about products, endorsements or other material information. |
| EU AI Act | Art. 50(2), (4) | These provisions establish role- and content-specific transparency duties, including provider marking of synthetic outputs and deployer disclosure for deepfakes, rather than a blanket labelling duty for all AI-assisted advertising. |
| GDPR | Arts. 5–6 | Where content generation processes identifiable people's images, voices or other personal data, brands must establish a lawful basis and comply with principles including purpose limitation, data minimisation and transparency. |
Sources
Court & Regulator Decisions
Infopaq International v Danske Dagblades Forening (opens in a new tab)
The originality standard and protection of short extracts inform checks for copied expression in AI-generated marketing, without establishing copyright protection for purely machine-generated outputs.
Kneschke v LAION (opens in a new tab)
The research-specific text-and-data-mining ruling helps distinguish dataset-building exceptions from permission to reproduce protected images in commercial marketing.
Thomson Reuters v Ross Intelligence (opens in a new tab)
The US ruling shows that AI training does not automatically qualify as fair use, informing scrutiny of providers’ claims about lawful source-material use.
Key Takeaways
- Check input permissions, output similarity and tool terms; do not assume AI-generated content is rights-free or exclusively ownable.
- Verify product claims, testimonials and endorsements before publication.
- Assess disclosure needs for synthetic content, especially realistic depictions that audiences could mistake for authentic evidence.
- Assign human approval responsibility and review content for stereotypes, exclusion and exploitation of vulnerable audiences.
Related Use Cases
From the Session
Marketing and Content in the Age of AI – Legal and Ethical Considerations
Speakers
