Copyright Protection Of Algorithmically Generated Swahili Children Rhymes.
1. Understanding the Issue
Algorithmically generated content raises key questions:
Authorship: Traditional copyright law protects works created by human authors. AI-generated works may not qualify unless there is human creativity involved in the input, prompts, or editing.
Originality: Copyright requires original expression, not mere ideas or facts.
Derivative works: Using pre-existing rhymes or datasets may implicate existing copyright.
Cultural expression: Swahili children rhymes may be traditional or folkloric. Traditional folklore is generally in the public domain, but new adaptations can be copyrighted.
For algorithmically generated rhymes:
If the AI is just a tool and a human selects, edits, or arranges the rhymes, copyright can apply to the human author’s contribution.
If fully autonomous, most jurisdictions may deny copyright protection since there is no human authorship.
2. Case Law Examples
Here are six relevant cases illustrating how courts handle authorship, originality, and derivative works, applicable to algorithmic rhymes:
Case 1: Feist Publications, Inc. v. Rural Telephone Service Co., 499 U.S. 340 (1991)
Facts: Feist copied phone directory listings.
Legal Principle: Facts are not copyrightable, only original expression is protected.
Application: Traditional Swahili rhymes that are folklore (public domain) cannot be copyrighted, but new, creative arrangements or algorithmic interpretations may be.
Case 2: Naruto v. Slater (Monkey Selfie Case), 2018, 4th Cir. 2018
Facts: A monkey took a selfie using a photographer’s camera. Who owns copyright?
Legal Principle: Only human authors can hold copyright; non-human creation is not eligible.
Application: Fully autonomous AI-generated rhymes without meaningful human input may not be copyrighted under U.S. law.
Case 3: Bridgeman Art Library v. Corel Corp., 36 F. Supp. 2d 191 (S.D.N.Y. 1999)
Facts: Photographs of public domain paintings were claimed as copyrighted.
Legal Principle: Exact reproductions of public domain works lack originality, so they are not protected.
Application: If an AI reproduces existing Swahili rhymes verbatim, copyright may not apply. Original creative variations are necessary.
Case 4: Burrow-Giles Lithographic Co. v. Sarony, 111 U.S. 53 (1884)
Facts: A photographer’s portrait was protected as original work.
Legal Principle: Copyright protects creative human authorship, not mechanical or automatic reproduction.
Application: Human selection and arrangement of algorithmically generated rhymes can qualify for copyright protection.
Case 5: Feist + Authors Guild v. Google (Google Books), 804 F.3d 202 (2d Cir. 2015)
Facts: Google scanned millions of books and made snippets.
Legal Principle: Automated copying can infringe if it reproduces original expression, but transformative uses may be fair use.
Application: AI-generated rhymes that transform public domain content may still be protected if humans creatively curated or modified them.
Case 6: Thaler v. Commissioner of Patents (Australia, 2022)
Facts: An AI system, “DABUS,” was named as inventor on patents.
Legal Principle: The court rejected non-human inventors; only humans can claim IP rights.
Application: AI cannot claim copyright. The human author or programmer who guides the AI or selects outputs is considered the author.
Case 7: Rogers v. Koons, 960 F.2d 301 (2d Cir. 1992)
Facts: Koons copied a photograph to make a sculpture.
Legal Principle: Derivative works must add originality, or they infringe.
Application: If algorithmic rhymes are generated from copyrighted children rhymes, you must transform or add creativity to avoid infringement.
3. Practical Implications
Human contribution is essential:
Selecting, editing, or curating the AI output can establish copyright ownership.
Originality matters:
Merely generating text via AI without creative input is generally not protected.
Public domain content:
Swahili folklore rhymes are free to use, but new expressions of them can be protected.
Derivative works caution:
Using copyrighted datasets or existing rhymes without transformation may lead to infringement.
4. Legal Strategy for VR/AI Content Creators
Document your creative process: Keep logs of human input, editing, and selection of AI outputs.
Transformative works: Add new musical arrangement, animation, rhythm, or narrative context.
Use public domain or licensed datasets: Avoid infringing existing copyrighted rhymes.
Consider moral and cultural rights: Even public domain folklore may require community permissions in some jurisdictions.
Summary Table of Cases & Lessons for AI-Generated Rhymes
| Case | Key Principle | Application to AI Rhymes |
|---|---|---|
| Feist v. Rural | Facts ≠ copyright | Folklore rhymes are free; original AI expressions can be copyrighted |
| Naruto v. Slater | Only humans can hold copyright | AI alone cannot own copyright |
| Bridgeman v. Corel | Exact reproductions lack originality | Simple copying of rhymes isn’t protected |
| Burrow-Giles | Human authorship required | Human curation of AI output enables copyright |
| Google Books | Transformative uses may be protected | AI output must show creative transformation |
| Thaler v. Commissioner | AI can’t be inventor | Humans must guide AI for IP rights |
| Rogers v. Koons | Derivative works require originality | AI adaptations of existing rhymes must be transformative |
RELATED Blog
comments