OwnershIP Concerns In Generative AI Used To Redesign Public Transport Routes.

18 Mar 2026 --
0 Comments

1. Ownership Concerns in Generative AI for Public Transport

When generative AI (like GPT-based systems, AI mapping tools, or route-optimization AI) is used to redesign public transport routes, several ownership and intellectual property (IP) issues emerge:

Authorship & Copyright
- AI generates new content (e.g., optimized routes, timetables, or network maps).
- The question: Who owns the output?
  - The AI developer?
  - The transport authority commissioning the work?
  - The AI itself? (legally currently impossible—AI cannot own IP).
Data Ownership & Input Rights
- AI requires training data, often from public transport schedules, GPS data, or commuter flow datasets.
- If proprietary data is used without permission, data copyright infringement can occur.
Derivative Work
- AI may produce routes that closely resemble existing ones.
- This could create derivative works, leading to potential infringement of the original data owner.
Patent Issues
- Novel AI algorithms for route optimization may be patented.
- Public transport authorities may face questions about licensing AI tools or algorithms.

2. Case Laws Relevant to AI Ownership and Copyright

Here are five key cases and legal discussions illustrating ownership concerns in AI-generated works:

Case 1: Naruto v. Slater (Monkey Selfie Case) – US, 2018

Facts: A macaque took a “selfie” using a photographer’s camera. The question was whether the monkey owned copyright.
Ruling: Court ruled animals (and by extension, AI) cannot own copyright.
Implication for AI in Transport:
- Any AI-generated transport route maps cannot be copyrighted by the AI itself.
- The human or entity directing the AI is usually considered the potential copyright owner.

Case 2: Thaler v. Commissioner of Patents – Australia, 2021

Facts: Stephen Thaler applied for a patent listing an AI as the inventor (the AI named DABUS).
Ruling: The Federal Court of Australia held that AI can be recognized as an inventor, but ownership still vests in the person who created or operates the AI.
Implication:
- For AI-generated transport optimizations, if AI contributes creatively, the operator may claim IP rights as the inventor/author.

Case 3: Authors Guild v. Google, US, 2015

Facts: Google scanned millions of books and created a searchable database. Authors claimed copyright infringement.
Ruling: Court sided with Google, recognizing transformative use as fair use.
Implication for AI in Transport:
- AI tools that use historical transport data to generate new optimized routes may qualify as transformative work, limiting infringement risk if outputs are significantly new and creative.

Case 4: Feist Publications v. Rural Telephone Service, US, 1991

Facts: The court ruled that mere compilations of facts (like phone directories) are not copyrightable unless they exhibit creativity.
Implication:
- AI-generated routes based solely on factual commuter data may not receive copyright protection unless they involve a creative selection or design (e.g., innovative optimization strategy).

Case 5: Naruto v. Slater-like AI scenario in EU (Case C-5/08 Infopaq International A/S v. Danske Dagblades Forening, 2009)

Facts: EU court emphasized that even small excerpts can be copyrighted if they show originality.
Implication for transport AI:
- If AI-generated route plans incorporate copyrighted planning strategies, even parts could be infringing.
- Careful attention is needed when AI “learns” from proprietary transit models.

Case 6: Warner Bros v. AI-Created Content – Hypothetical / Emerging

Context: AI models trained on copyrighted media produce derivative works.
Implication:
- For public transport, if AI uses proprietary route optimization algorithms or mapping data, agencies must license data properly.
- Raises data ownership concerns, even if the AI generates a new map.

3. Key Takeaways for Public Transport Authorities

Ownership defaults to the human/operator who manages the AI, not the AI itself.
Ensure licensing of input data – both public and private datasets.
Protect innovative outputs – if AI-generated routes are highly creative, agencies could potentially claim copyright or patent protection.
Watch for derivative risks – AI output may resemble existing protected routes, creating infringement issues.
Document human input – courts often look at the extent of human creative control to determine ownership.

4. Conclusion

Using generative AI for public transport redesign involves a complex mix of copyright, patent, and data ownership issues. The main points from case law are:

AI cannot hold IP rights itself (Naruto v. Slater).
Human operators may be considered authors or inventors (Thaler v. Commissioner of Patents).
Transformative or creative outputs may be protected or permissible under fair use (Authors Guild v. Google).
Fact-based outputs alone may not be copyrightable (Feist v. Rural).
Licensing of underlying data and careful human oversight is essential.