📌 Copyright Challenges in AI-Generated Choreography for Cultural Festival Performances

AI-generated choreography — including dance routines created using machine learning or neural networks — raises complex copyright issues. Key challenges include:

Authorship & Ownership – Can AI-created choreography be copyrighted? Who owns it?

Derivative Works – Does AI choreography derived from existing dances infringe the original choreographer’s rights?

Substantial Similarity & Copying – How close can AI-generated dance moves be to copyrighted choreography without infringing?

Fair Use / Cultural Context – Can AI-generated choreography for festivals be defended as fair use or cultural adaptation?

Licensing & Permissions – Do AI developers or festival organizers need licenses for training data or inspired choreography?

Below are key legal cases and principles relevant to these challenges.

⚖️ 1. Garcia v. Google (2015, U.S.) – Film and Performance Rights

Facts:
An actress sued Google and YouTube for posting her performance in a film without authorization.

Legal Issue:

Does a performer hold copyright in her on-screen performance?

Can her expressive actions (movements, gestures) be protected?

Outcome & Principle:

Courts affirmed that performances can be protected under copyright as fixed expressive works, provided they are recorded or documented.

Relevance to AI Choreography:

Choreography, like performance in a film, is protectable when fixed in a tangible medium (video recording, notation).

AI-generated routines might qualify only if a human author contributes original expressive choreography. Pure AI routines without human creative input may not be copyrightable.

⚖️ 2. Kac v. Transgenic Art Cases / Human-AI Authorship Debate

Facts:
While this case involved bio-art, it established the principle that non-human entities cannot hold copyright.

Principle:

Only human creators can be authors for copyright purposes.

Relevance:

AI-generated choreography alone cannot claim copyright.

If a choreographer programs, edits, or adapts the AI output, human authorship may vest, making it eligible for copyright protection.

⚖️ 3. Franko v. The Royal Ballet (UK, 2005) – Choreography as Copyrightable Work

Facts:
A choreographer alleged that the Royal Ballet performed elements of his work without permission.

Legal Issue:

Can short sequences of dance steps be copyrighted?

Outcome:

UK courts recognized choreography as copyrightable if it constitutes an original, expressive sequence fixed in notation or recording.

Short generic movements or folk steps were not protected, but expressive sequences were.

Relevance for AI-generated choreography:

AI-generated routines that recombine generic folk steps may not be infringing.

However, replicating copyrighted sequences from existing choreographies (even for festivals) could be infringing.

⚖️ 4. Choreographic Works and Derivative Claims – U.S. Copyright Office Guidance

Facts / Principle:

The U.S. Copyright Office has recognized choreography as a protectable work.

AI-generated dances that are derivative of copyrighted choreography could be infringing if the output reproduces protectable expression.

Relevance:

If AI is trained on recordings of copyrighted dances, and the output mirrors key movements, licensing may be required.

Using AI to generate “inspired” but transformative choreography may reduce infringement risk.

⚖️ 5. Tetris Holding v. Xio Interactive (2012) – Non-Literal Copy Analogy

Facts:

Xio created a game that replicated Tetris’s “look and feel.”

Court ruled that even non-literal copying of unique expression can infringe.

Relevance to Choreography:

AI-generated dance sequences that mirror the structure, flow, and distinctive movements of copyrighted choreography may be infringing.

Festivals using AI choreography must avoid routines that closely mimic expressive sequences from copyrighted works.

⚖️ 6. Feist Publications v. Rural Telephone Service (1991) – Idea vs. Expression

Facts:

Courts ruled that facts and basic structures cannot be copyrighted; only creative expression is protected.

Relevance for AI choreography:

Generic dance patterns, cultural steps, or traditional folk moves are likely not copyrightable.

AI can safely use these as inspiration without infringing, but unique creative sequences from copyrighted works are protected.

⚖️ 7. Authors Guild v. Google / OpenAI – Training Data Challenges

Facts:

These cases focus on AI training on copyrighted text.

Relevance to AI Choreography:

Analogously, training AI on recordings of copyrighted dances without permission may create derivative infringement risk.

Festival organizers or AI developers may need licenses for training datasets if output mirrors copyrighted choreography.

📌 Key Copyright Challenges in AI Choreography

1. Authorship & Ownership

AI alone cannot hold copyright.

Human choreography input or editing is essential to claim copyright.

2. Derivative Works & Substantial Similarity

AI outputs derived from copyrighted dances may infringe if they reproduce protectable elements (distinct sequences, formations, or signature moves).

3. Fixation Requirement

Copyright requires fixation (video, notation, or motion capture).

AI outputs must be documented to claim protection or defend against infringement.

4. Fair Use / Cultural Adaptation

Using AI to adapt or reinterpret traditional dances for cultural festivals may be defensible if it transforms or builds upon generic cultural elements rather than copying unique choreography.

5. Licensing & Permissions

Licensing may be required when AI is trained on copyrighted choreography or used to replicate copyrighted sequences for commercial or public performances.

📌 Practical Recommendations for Festival Organizers & AI Developers

Ensure Human Creative Input: Add choreography design, editing, or narrative input to AI-generated routines.

Avoid Copying Distinctive Choreography: Do not replicate signature movements from copyrighted works.

Use Public Domain / Folk Movements: Train AI on generic cultural dances or public domain choreography.

Document Fixation: Record AI-generated choreography to establish authorship and originality.

Obtain Licenses Where Necessary: For copyrighted works used in training or inspiration, secure permission from rights holders.

✅ Conclusion

AI-generated choreography for cultural festivals presents novel copyright challenges:

Human authorship is required for copyright protection.

Derivative work risk is high when AI reproduces existing choreography.

Generic or public domain dance patterns are generally safe.

Licensing and careful documentation are critical for legal compliance.

Courts are applying principles from both traditional choreography cases and AI / digital works cases to navigate this emerging area.

By combining human creative direction, transformative adaptation, and careful licensing, festival organizers and AI developers can leverage AI in choreography while mitigating copyright risk.