Contact Us

Copyright and AI-Generated Content: Your Marketing Team's Guide

US/EU/Asia law, platform licenses, and protection strategies explained

Purely AI-generated content cannot be copyrighted in the US or EU.

The Bottom Line

Your competitors can legally copy AI-generated marketing assets that lack human creative input. The March 2025 Thaler v. Perlmutter ruling confirmed copyright requires human authorship, while Anthropic's $1.5 billion settlement proved AI companies face massive liability for using pirated training data. Bottom line: treat AI as your assistant, not your creator.

What Gets Copyright Protection?

The US Copyright Office's January 2025 report is clear: prompts alone, even hundreds of iterations, don't establish authorship because you cannot predict or control AI output.

Thaler v. Perlmutter (March 2025) confirmed machines cannot be authors. Zarya of the Dawn established the boundary: human-written text and creative arrangement got copyright protection, but individual Midjourney images did not.

Protected

Human-authored text and arrangement

Substantial human modifications

AI-assisted works where humans control expression

Not Protected

Raw AI outputs

Prompt refinement alone

Prompt refinement alone

EU Law and Major Legal Developments

The EU requires works to be "the author's own intellectual creation." AI outputs alone don't qualify. The EU AI Act (effective August 2026) mandates AI-generated content be labeled as such. The GEMA v. OpenAI ruling (November 2025) in Munich found ChatGPT infringed copyright by reproducing song lyrics. OpenAI must pay licensing fees.

Critical US cases: The New York Times v. OpenAI tests whether AI training constitutes fair use. Thomson Reuters v. Ross Intelligence (February 2025) rejected fair use for AI training. The first major ruling established AI training on copyrighted content as commercial, non-transformative use causing market harm. Bartz v. Anthropic's $1.5 billion settlement (September 2025) proved using pirated training data is never protected, even if AI methodology might be fair use.

Platform Rights: Know Before You Create

Marketing teams can use AI content commercially, but protections vary dramatically:

Platform Commercial Use Indemnification Critical Notes
OpenAI
(ChatGPT/DALL-E)
 ✅ All paid plans ✅ Enterprise/API only
(up to 12 months fees)		 Free tier gets rights but zero protection
Google Gemini
 ✅ Free & paid ✅ Paid plans only Free tier: Google uses your prompts for training. Paid: full indemnification, no training on your data
Midjourney
 ✅ Paid only ❌ None Free tier is non-commercial. Images are public by default
Adobe Firefly
 ✅ Yes ✅ Enterprise
($10K cap)		 "Commercially safe" training. Auto-tags as AI
Stability AI
 ✅ Under $1M revenue ❌ None Enterprise license required above $1M
FAQ

Not if it’s purely AI-generated. US and EU law require human authorship for copyright protection.

Thaler v. Perlmutter (March 2025): DC Circuit Court ruled machines cannot be authors under US copyright law.

Zarya of the Dawn (2023): Copyright Office granted protection for human-written text and creative arrangement but denied it for raw Midjourney images.

What gets copyright protection:

  • Human-authored text and strategic direction
  • Creative selection and arrangement of AI elements
  • Substantial human modifications to AI outputs
  • Works where humans control creative expression

What doesn’t get protection:

  • Raw AI outputs from prompts
  • Minor adjustments to AI content
  • Iterative prompt refinement alone
  • Content where AI determined creative choices

For copyright protection, human creative contribution must be substantial and documented.

Ownership depends on contracts and the level of human contribution.

Most AI platforms grant users contractual rights to outputs, but copyright protection depends on whether human contribution meets legal thresholds for authorship.

Employment context:

  • Content created by employees: employer owns (work-for-hire doctrine)
  • Content created by contractors: ownership determined by contract
  • Content created by agencies: requires explicit transfer agreement

Always verify:

  • Platform commercial use terms
  • Whether the platform trains on your data
  • Indemnification coverage (if any)
  • Geographic restrictions on usage

The user bears liability, not the AI platform. Platform indemnification (where offered) provides limited protection.

Platform indemnification examples:

  • OpenAI Enterprise: Copyright Shield covers claims up to 12 months of fees paid (Enterprise/API only, not free tier)
  • Google Gemini paid plans: Covers both training data and generated outputs (free tier excluded)
  • Adobe Firefly Enterprise: Indemnification capped at $10,000 per output or claim
  • Midjourney, Stability AI: No indemnification offered

 

What indemnification typically doesn’t cover:

  • Trademark claims (explicitly excluded)
  • Content you knew or should have known was infringing
  • Outputs where you disabled safety filters or citations
  • Use after receiving infringement notice from rights holder

 

Legal context: The $1.5 billion Anthropic settlement (September 2025) involved authors proving Anthropic downloaded pirated books for training data. Documentation of human creative contribution is your primary defense.

Depends on jurisdiction and content type.

United States (current):

  • No general legal requirement for marketing content
  • FTC requires disclosure if AI creates deceptive endorsements, testimonials, or product claims
  • State laws emerging (California, New York considering disclosure mandates)

 

European Union (August 2026):

  • EU AI Act mandates labeling for AI-generated images, audio, and video
  • Required format: “clearly labelled as AI generated” in machine-readable means
  • Text content currently exempt
  • Penalties for non-compliance: up to €35M or 7% global annual revenue

 

Platform-specific requirements:

  • TikTok, Instagram, YouTube adding AI content labels
  • Some platforms require creator disclosure when using AI generation tools

Transparency requirements are tightening globally. Plan for disclosure capabilities even where not currently required.

Free tools provide no legal protection. Paid enterprise platforms offer indemnification, privacy, and clear commercial rights.

Legal FeatureFree ToolsEnterprise Platforms
Copyright indemnification❌ None✅ Available
Data privacy❌ May train on your data✅ No training on customer data
Commercial use rights⚠️ Often restricted✅ Unrestricted
Usage tracking❌ Prompts may be stored/analyzed✅ Private

Real risk: Midjourney free tier restricts commercial use entirely (Creative Commons Noncommercial 4.0 license). Using free-tier outputs commercially violates license terms.

Enterprise platform pricing typically ranges from $30-60 per user monthly.

Yes, if content is purely AI-generated without substantial human creative contribution.

Legal principle: Pure AI outputs lack copyright protection under current US and EU law. Content in the public domain can be copied freely.

What creates legal protection:

  • Substantial human creative direction
  • Selection and arrangement decisions
  • Editing and modification choices
  • Strategic composition

The Zarya of the Dawn case established this: individual AI images received no copyright protection, but human-authored arrangement and text did.

Proving “substantial human contribution” requires documentation. Without evidence of the creative process, defending against copying becomes difficult.

Trademark infringement is serious exposure. Platform indemnification explicitly excludes trademark claims.

The legal distinction:

  • Copyright protects creative expression
  • Trademark protects brand identifiers and prevents consumer confusion

 

Common AI trademark risks:

  • Visual similarity to protected logos, packaging, or designs (Nike swoosh, Apple aesthetic, luxury brand styling)
  • Trademarked color combinations (UPS brown, Tiffany blue, T-Mobile magenta)
  • Trade dress (overall look and feel of products or packaging)

 

Why indemnification doesn’t cover trademarks: The infringement occurs through your commercial use creating consumer confusion, not through AI generation itself. This is your liability.

Recent precedent: GEMA v. OpenAI (Munich, November 2025) found ChatGPT infringed copyright by reproducing song lyrics. OpenAI must pay licensing fees.

Risk management:

  • Avoid prompts referencing competitor brands by name
  • Screen outputs for brand resemblance before commercial use
  • Reverse image search all visual content
  • Legal review for major campaigns

Trademark litigation is expensive. Most platform indemnification won’t help.

Style and aesthetic aren’t copyrightable. Specific creative expressions are.

Legal (acceptable prompts):

  • “Luxury automotive photography aesthetic”
  • “Minimalist tech product design language”
  • “Premium spirits brand sophistication”
  • “High-energy sports marketing vibe”

 

Illegal (infringement risk):

  • “Apple iPhone campaign clone”
  • “BMW 2024 ad series recreation”
  • “Patagonia campaign exact match”
  • Prompts naming specific artists, photographers, or brands

 

Key legal case: Thomson Reuters v. Ross Intelligence (February 2025) rejected fair use defense for AI training on competitor’s content. Court found it was commercial, non-transformative use creating a competing product causing market harm.

Legal principle: You can capture quality level and feeling. You cannot copy specific protected elements.

The line: “inspired by” is generally acceptable. “Identical to” crosses into infringement.

Documentation proves human authorship and protects against infringement claims.

Essential records to maintain:

Creative development:

  • Creative briefs and strategic direction
  • Mood boards and reference materials
  • Initial concept development

 

Production records:

  • Prompt history and iterations
  • Raw AI outputs before editing
  • Human modifications and editing decisions
  • Version history with timestamps

 

Verification:

  • Reverse image search results
  • Plagiarism check reports (for text content)
  • Trademark screening results
  • Brand similarity reviews

 

Approval chain:

  • Quality review notes
  • Stakeholder approvals
  • Final sign-offs with dates

 

Platform documentation:

  • Commercial license confirmations
  • Terms of service at time of creation
  • Indemnification coverage details

 

Why this matters: The $1.5 billion Anthropic settlement turned on evidence. Anthropic couldn’t prove legitimate training data sourcing. Documentation is legal defense.

Retention period: Minimum 7 years (statute of limitations for copyright claims in most jurisdictions).

Store in version-controlled system with audit trail. Cloud storage with automated backups recommended.

Yes, when you understand the legal requirements and limitations.

Legal requirements by scale:

Small-scale use (under 100 assets/month):

  • Free tools with commercial licenses acceptable
  • Document human contributions
  • Basic verification workflows

 

Medium-scale use (100-1,000 assets/month):

  • Enterprise platforms with indemnification recommended
  • Systematic documentation processes
  • Regular legal review of outputs
  • Trademark screening protocols

 

Large-scale production (1,000+ assets/month):

  • Enterprise platforms mandatory
  • Comprehensive documentation systems
  • Dedicated legal review workflows
  • Indemnification coverage essential

 

By client type:

Agencies: You’re liable to clients for delivered content. Platform indemnification doesn’t transfer automatically. Contracts must specify AI usage, liability allocation, and deliverable ownership.

Brands: Focus on protecting your own trademark while avoiding others. FTC holds you responsible for AI-generated product claims regardless of source.

Manufacturers/Retailers: E-commerce at scale needs consistent processes. One infringement claim across thousands of SKUs can shut down entire catalogs.

Recent legal developments:

  • March 2025 – Thaler v. Perlmutter: Established human authorship requirement
  • September 2025 – Anthropic settlement: $1.5 billion for using pirated training data
  • November 2025 – GEMA v. OpenAI: First EU finding of AI model copyright infringement

Risk comes from treating AI carelessly: using free tools commercially without proper licenses, lacking documentation, no verification processes, unclear liability allocation.

Copyright principles align globally (human authorship required), but platform availability and enforcement vary significantly.

Platform availability by region:

China:

  • OpenAI (ChatGPT, DALL-E) blocked
  • Google services restricted
  • Domestic alternatives: Alibaba Tongyi Qianwen, Baidu ERNIE, ByteDance platforms
  • Commercial licenses required from Chinese platforms

 

Hong Kong, Singapore, Japan, South Korea:

  • Full access to Western enterprise platforms
  • Strong IP protection frameworks
  • Reliable enforcement mechanisms

 

Copyright law variations:

China:

  • Copyright Law (2020 revision) requires human authorship
  • Registration with National Copyright Administration recommended for enforcement
  • First-to-file system means documentation timing matters
  • Cross-border enforcement through WIPO treaties but practically challenging

Japan:

  • Strong copyright protection similar to US/EU frameworks
  • AI-generated content follows human authorship requirement
  • Registration optional but provides litigation advantages

Singapore:

  • Common law system aligned with UK/US principles
  • Robust IP enforcement
  • Regional hub for APAC content production

 

EU AI Act considerations for exports:

If you’re creating content in US/Asia for EU markets, August 2026 labeling requirements apply:

  • Images, audio, video must be labeled as AI-generated
  • Content published or distributed in EU
  • Penalties for non-compliance up to €35M or 7% global revenue

 

Multi-jurisdiction campaigns require:

  • Platform selection based on regional availability
  • Jurisdiction-specific copyright documentation
  • Compliance with strictest applicable law (often EU)

Running campaigns across US, EU, and China requires different platform strategies, documentation approaches, and legal compliance frameworks for each region.

You Own the Liability

Platform terms shift infringement liability to users. "The AI did it" provides no defense. Marketing teams face exposure for:

  • Copyright infringement: Using content that mimics protected works
  • Trademark violations: AI-generated content resembling protected brands (typically excluded from indemnification)
  • False advertising: Unverified AI-generated product claims violate FTC rules
  • Defamation and privacy: AI producing false statements or incorporating personal data

Protection Strategy

Since purely AI-generated content may be unprotected, competitors can copy your assets. Build protection through deliberate human involvement:

  • Layer human creativity: Use AI for drafts, apply significant human editing and creative judgment
  • Document everything: Archive prompts, raw outputs, edits, and final versions with timestamps
  • Avoid mimicry: Use descriptive prompts, not references to specific artists or brands
  • Verify outputs: Reverse image search visuals, plagiarism-check text, review for likenesses
  • Require human approval: Legal review for high-stakes campaigns before publication

What This Means for Your Marketing Team

Use AI confidently with appropriate safeguards. Human authorship remains the threshold for copyright protection in both the US and EU. Purely AI-generated content offers no competitive moat.

The $1.5 billion Anthropic settlement and GEMA v. OpenAI ruling prove courts will impose substantial consequences for copyright violations. Organizations establishing rigorous AI governance now will be best positioned as this legal landscape evolves.

Your action plan: Choose enterprise-tier platforms with indemnification (OpenAI Enterprise, Google Gemini paid plans, Adobe Enterprise). Document human contributions. Implement verification workflows. Treat AI as a powerful tool, not an autonomous content department.

IMPORTANT DISCLAIMER

AI copyright law is evolving rapidly. Courts worldwide are actively deciding cases that will shape how intellectual property rights apply to AI-generated content. New legislation, regulatory guidance, and judicial precedents emerge regularly.

 

This content is provided for educational purposes only and does not constitute legal advice. While we strive for accuracy, we do not guarantee that all information on this page is current, complete, or applicable to your specific situation. Laws, platform terms, and enforcement practices may have changed since publication.

Before making business decisions based on this information:

  • Consult qualified legal counsel in your jurisdiction
  • Verify current platform terms and conditions
  • Review the latest regulatory requirements for your markets
  • Assess your specific risk profile and compliance needs

The legal landscape for AI-generated content differs significantly across jurisdictions and continues to develop. What is legally permissible today may change tomorrow.

Last updated: December 2025