Ownership Of Cognitive Ai Systems Generating New Logical Languages For Programming.
1. Overview: AI-Generated Programming Languages
When a Cognitive AI system generates a new logical language (or a programming language), several legal questions arise:
Authorship: Who owns the creation—the AI, the programmer, or the organization deploying the AI?
Intellectual Property Protection: Can copyright, patents, or trade secrets protect an AI-generated language?
Legal Precedents: Courts have increasingly examined AI-generated works, especially after cases involving AI-authored content (e.g., music, code, art).
Two frameworks are generally considered:
Copyright Law: Protects original works of authorship fixed in a tangible medium (e.g., code, text).
Key question: Can an AI system be an “author”? Most courts say no; a human must contribute significant creative input.
Patent Law: Protects novel and non-obvious inventions.
Key question: Can an AI system be an “inventor”? Courts currently hold that only humans can be inventors.
Trade Secret Law: Protects proprietary methods or algorithms not publicly disclosed.
Ownership is usually held by the company that owns the system generating the language.
2. Case Law Examples
Case 1: Naruto v. Slater (2018) – Copyright and Non-Human Authorship
Facts: A monkey took a selfie, and the copyright office initially rejected copyright claims for non-human authorship.
Holding: Copyright cannot be granted to non-human entities.
Implication for AI-generated programming languages:
If a cognitive AI creates a language independently, it cannot hold copyright. Ownership defaults to the human or organization responsible for training or operating the AI.
Case 2: Thaler v. Commissioner of Patents (DABUS AI Cases, US & UK, 2021–2022)
Facts: Stephen Thaler listed DABUS AI as the inventor of new inventions. The applications claimed that AI could be an inventor.
Holding:
UK: Patents must name a human inventor. AI cannot be listed.
US: Patent office rejected the application for the same reason.
Implication: Even if AI generates a novel programming language or logical system, patent rights can only belong to a human inventor, not AI.
Case 3: Apple v. Samsung (2012) – Software and Algorithm Ownership
Facts: Apple sued Samsung for copying iOS design elements.
Holding: Software code, design, and user interface elements can be copyrighted.
Implication:
If an AI develops a new programming language, the code implementing it could potentially be copyrighted—but only if a human significantly contributes or supervises its creation.
Case 4: Feist Publications v. Rural Telephone Service (1991) – Originality Requirement
Facts: Rural Telephone had a phone directory; Feist copied it. Rural argued that the compilation was copyrightable.
Holding: A compilation must have minimal creativity; mere facts (like phone numbers) aren’t protected.
Implication:
An AI generating a logical language must involve human creativity in its design to qualify for copyright. If the AI autonomously generates syntax or semantics, courts may deny copyright protection.
Case 5: Microsoft v. AT&T (2007) – Software and International Copyright
Facts: Microsoft distributed software globally; AT&T alleged patent/copyright infringement.
Holding: Copyright covers software code as a fixed expression; patent claims are separate.
Implication:
If an AI creates code that embodies a new logical language, copyright may apply to the implementation, but patent protection would require a human inventor.
Case 6: Naruto v. Slater Extension: AI Code Generators
Fact Pattern Analogy: Modern AI code generators like GitHub Copilot produce code based on human prompts.
Legal Analysis: Courts suggest that ownership may lie with the human giving creative instructions, or the company hosting the AI, rather than the AI itself.
3. Key Legal Principles for Ownership of AI-Generated Languages
Human Authorship is Essential for Copyright.
Courts consistently rule that non-human entities (AI or animals) cannot hold copyrights.
Patents Require Human Inventorship.
AI cannot be legally recognized as an inventor, so any patentable features of a new logical language must involve human creativity.
Corporate Ownership via Work-for-Hire.
If an AI is developed by a company and used to create a programming language, the company typically owns the rights under work-for-hire or employee-invention doctrines.
Trade Secrets Are Flexible.
Proprietary AI-generated programming languages may be protected as trade secrets, independent of human authorship or inventorship.
4. Practical Implications
For AI Developers: Document human involvement in the creation of logical structures or syntax to ensure copyright/patent protection.
For Companies: Ownership is safest when AI outputs are clearly generated under company supervision.
For Innovators: Trade secret protection is often the most robust option for purely AI-generated code or logical languages.
Summary Table of Case Implications:
| Case | Year | Principle | Implication for AI Programming Language |
|---|---|---|---|
| Naruto v. Slater | 2018 | Non-humans cannot hold copyright | AI alone cannot be author |
| Thaler v. DABUS | 2021 | Only humans can be inventors | AI cannot hold patent rights |
| Apple v. Samsung | 2012 | Software is copyrightable | Human-assisted AI code can be protected |
| Feist v. Rural | 1991 | Originality required | AI-only creativity may fail copyright |
| Microsoft v. AT&T | 2007 | Software implementation copyright | Implementation of AI language may be protected |
| GitHub Copilot Analogy | 2020s | Human input matters | Human prompt ownership critical |
Conclusion
Ownership of AI-generated programming languages is legally complex:
Copyright: Only humans can hold rights. AI-only output is not protected.
Patent: Human inventorship is required.
Trade Secret: Company ownership is possible without human authorship.
This creates a practical strategy: to claim ownership, involve humans in design, supervision, or implementation, and maintain internal records documenting this involvement.
RELATED Blog
comments