dub-domain-verification=c7e774f0-eff0-40da-a7a2-22a7edfd75a8
top of page
Search

AI and Branding 2026: Copyright Risks for Content Creators

  • Writer: Harriet Moser
    Harriet Moser
  • Jan 5
  • 7 min read

You see them everywhere on social media: AI-generated images of celebrities in videos they never consented to. Disney and Pixar characters in content they never approved. Brand logos seamlessly integrated into campaigns without permission.


The uncomfortable truth? Just because AI tools allow you to create such content doesn't mean you should - or legally can.


AI and Branding In 2026

Before diving into the copyright issue, we need more context. AI has evolved from experimental novelty to strategic imperative in brand management. Industry analysts report that over 70% of companies are already achieving ROI from generative AI (Grannis, 2025).


Leading brands are already successfully deploying AI across diverse areas, such as:

  1. Strategic Analysis:

    AI processes market data, competitive intelligence, and consumer sentiment

  2. Brand Perception Monitoring:

    Real-time sentiment analysis identifies gaps between brand identity and public perception

  3. Visual Content Generation:

    Tools like Midjourney, DALL-E, and Adobe Firefly democratise high-quality visual creation

  4. Content Strategy Optimisation:

    AI-powered SEO tools enable smaller organisations to compete with enterprise operations

  5. Multi-Channel Personalisation:

    AI adapts brand communication across touchpoints in an automated manner


The Critical AI Copyright Risk Question

The AI copyright risk debate operates on two distinct but interconnected fronts:


Front 1: AI Providers vs Rights Holders

Over 50 active lawsuits in the United States alone challenge whether AI companies can legally train models on copyrighted books, articles, images, and music without permission or payment (Copyright Alliance, 2025). Authors, journalists, musicians, and visual artists argue that ingesting their works - even for "transformative" purposes—constitutes copyright infringement.

Courts show mixed signals. Some judges find AI training "extraordinarily transformative" and therefore fair use; others emphasise market harm and reject this defence. The U.S. Copyright Office's May 2025 report concludes pragmatically: "Some uses of copyrighted works for generative AI training will qualify as fair use, and some will not" (U.S. Copyright Office, 2025, siehe auch Skadden, 2025). Experts predict no definitive clarification until at least mid-to-late 2026 (MIT Technology Review, 2025).


Front 2: Creators Using AI Outputs

This is where it becomes personal. When someone uses AI to create an image of a celebrity promoting their product or generates a Disney character for marketing, they - not necessarily the AI provider - bear the legal consequences. This distinction is critical: training data lawsuits target AI companies' practices, but output liability also affects individuals generating AI content (Library of Congress, 2025).


Regional Legal Landscapes

Schweiz and Europe: Structured Regulation

The EU AI Act (applicable from August 2026) establishes the world's first comprehensive AI regulatory framework (European Union, 2024). For brand managers and content creators, several provisions carry immediate practical implications:

  • Transparency Requirements: AI providers must publish training data summaries and establish complaint procedures (Maienza & Garufi, 2025)

  • Copyright Compliance: Providers must respect EU copyright law and text/data mining opt-outs

  • Content Labelling: AI-generated content must be labelled (December 2025 draft Code of Practice) (European Commission, 2025)


Legal scholars such as Guadamuz (2024) criticise that the legislation provides minimal protection for creators whose works were already scraped before August 2025—effectively granting AI companies "a free lunch" for historical infringements (Martens, 2025).


Switzerland has chosen a distinctly measured path (Swiss Federal Council, 2025). DThe Gössi Motion (March 2025) demands that rights holders must give explicit permission before their works can be used to train AI systems (Swiss Federal Institute of Intellectual Property, 2025). Implementation legislation is expected by the end of 2026. Switzerland plans to ratify the Council of Europe's AI Convention (Council of Europe, 2024) and pursue a sector-specific approach.


For Swiss businesses, the practical reality is dual compliance: respecting emerging Swiss requirements whilst aligning with EU standards. Swiss brands operating across borders must adhere to the stricter standard.


United States: Fair Use Battles

Three court decisions from 2025 illustrate conflicting interpretations—legal experts emphasise the inconsistency of these rulings (Smalley & Palfreyman, 2025):

  • Bartz v. Anthropic (Juni 2025) (Bartz v. Anthropic PBC, 2025):  Judge Alsup ruled that AI training is "extraordinarily transformative" and fair use, but rejected the use of pirated copies

  • Kadrey v. Meta (März 2025) (Kadrey v. Meta Platforms, 2025):  Found fair use but noted the ruling doesn't mean Meta's use is lawful—plaintiffs simply didn't prove their case

  • Thomson Reuters v. ROSS Intelligence (Februar 2025) (Thomson Reuters v. ROSS Intelligence, 2025): Rejected fair use because the AI served the same market function as the original work


Whilst copyright law remains unclear, personality rights offer clearer grounds. California's AB 2602 and AB 1836 (effective January 2025) (Hendrickson, 2025) expressly prohibit unauthorised use of digital replicas, including AI-generated likenesses.


The Midler v. Ford case (1988) (Poler, 2023) established that celebrities have protectable property interests in their voices and likenesses—a precedent that applies directly to AI-generated deepfakes.


Asia: A Fragmented Landscape

Japan:  Article 30-4 of the 2018 Copyright Act amendments (Japan Agency for Cultural Affairs, 2024) allows broad use of copyrighted works for AI training without permission, even commercially, positioning Japan as an attractive location for AI development (Mitsui & Co., 2025). However, pushback from the creative industries has prompted guidelines clarifying limits.


Singapore:  The 2021 Copyright Act amendment (Hays, 2024) permits works for "computational data analysis" with lawful access—positioning Singapore as an AI hub.


China:  According to Hays (2024), China takes a distinctive "moderately lenient" approach—lenient towards AI training whilst imposing strict controls on generated content. The Interim Measures for the Management of Generative AI Services (August 2023) require providers to "respect intellectual property rights" and implement content moderation, but courts have not ruled that AI training inherently infringes copyright. Mandatory content labelling (effective September 2025) requires explicit watermarks or implicit metadata for all AI-generated content (Hays, 2024).


The Unsettling Truth About Viral AI Content

Every AI-generated image using protected elements potentially violates:

  • Personality rights (for individuals' likenesses)

  • Trademark law (for brand identifiers)

  • Copyright law (for creative works and characters)

  • False endorsement statutes


The fact that "everyone is doing it" provides no legal defence. Rights holders increasingly monitor AI-generated content, and enforcement actions are accelerating.


Personal Perspective

I find viral AI content featuring unauthorised celebrity likenesses and brand logos genuinely unsettling—not from legal paranoia, but from an ethical standpoint. These images represent creativity without consent, profit without permission, and engagement built on deception. The ease of image generation doesn't diminish the wrongfulness.


Who Bears the Risk?

Legal liability operates across multiple levels, with individual creators bearing primary risk:

  1. Direct Infringement: You actively create and distribute the infringing content

  2. Easier Target: Pursuing individuals is simpler than litigating against financially strong AI companies

  3. AI Provider Disclaimers: AI providers' terms of use expressly disclaim liability for user content

  4. Evidence Clarity: Published infringement is straightforward to identify

  5. Statutory Damages: Up to USD 150,000 per work for wilful infringement


The Way Forward

The core message bears repeating: Just because AI tools enable you to create content featuring celebrities, brand logos, or copyrighted characters doesn't mean you should.


For Switzerland and European creators, the regulatory framework is crystallising towards transparency, accountability, and creator protection. The EU AI Act and pending Swiss copyright reforms establish clear expectations: respect intellectual property, secure proper permissions, and maintain human oversight.


My Recommendation:

  1. Invest in properly licensed AI tools and original content creation

  2. Maintain human oversight and creative direction

  3. Develop distinctive brand identities rather than imitating others

  4. Communicate transparently about AI use

  5. Respect intellectual property rights as a fundamental ethical standard


AI is transforming branding—but transformation doesn't mean abandoning legal and ethical principles. The most sustainable path forward pairs technological innovation with genuine respect for human creativity and intellectual property rights.


References


---

 

About the Author

Harriet Moser is an AI expert specialising in branding and founder of Ask Harriet, a human-led AI training consultancy based in Switzerland. As part of her master's thesis, she developed the AI Brand Strategy Framework for integrating artificial intelligence across five levels of brand management. She specialises in supporting companies and individuals in using AI systematically, ethically, and effectively.

For more information: www.askharriet.ch | LinkedIn: linkedin.com/in/harriet-moser

 

---

 

DISCLAIMER: I am not a lawyer. This article is for educational and informational purposes only. Whilst every effort has been made to ensure accuracy at the time of publication (January 2026), laws regarding AI and copyright are rapidly evolving. This article should therefore not be construed as legal advice. For specific legal guidance, please consult a qualified legal professional.


Blue bird on scales of justice: AI content vs. consent

 
 
bottom of page