GPT-4o: The idea of using tokenization to copyright a writing style is a forward-thinking approach that leverages technological advancements to address emerging challenges in IP protection

Question for AI Services: This news item titled “Troy professor copyrights creative writing style” indicates that Business Professor Dr. Rodger Morrison now has a copyright for a tokenisation regarding artificial intelligence. Professor Morrison’s comments are: “But if you can protect the style of a work, then no one can replicate your work which makes the work that YOU create much more valuable. And again it could be text, it could be audio samples it could be anything. Protect that. Now I have a peace of mind. I can protect my work. I am who I am. I might still be a horrible writer, but I’m a protected horrible writer.” The reporter states: “Tokenisation refers to the process of converting the text you put into things like ChatGPT into something smaller like tokens or subwords. Well, the tokens are used to then train AI to emulate specific styles to replicate the creative works of people to create things like poems. Dr. Morrison figured out how to create a tokenisation for his unique writing style, displayed in a book he wrote in 2002 called ‘The My Dearest Letters’. He was then awarded a copyright for that style, making this a huge advancement for other creators to protect their Intellectual Property Rights within the evolving world of Artificial Intelligence’. Explain whether you feel this can indeed be a technologically and legally sound approach for authors and whether you are aware of any pros and cons or caveats. Also, please do suggest improved, additional, or alternative means of achieving the same goal, whether technical remedies or not.

ChatGPT-4o’s answer:
The concept of copyrighting a creative writing style through tokenization, as described by Dr. Rodger Morrison, is intriguing and raises important questions about the intersection of technology and intellectual property law. Let’s explore the technological and legal soundness of this approach, as well as its potential pros, cons, and alternatives.

Technological Soundness

Pros:

1. Unique Identification: Tokenization can uniquely identify and capture the essence of an author’s style. This involves breaking down text into smaller units, like tokens or subwords, which can then be used to train AI models to recognize and reproduce specific writing styles.

2. AI Training: Using these tokens to train AI can help prevent the replication of an author's style without permission. This could be particularly useful in preventing plagiarism and ensuring original content remains identifiable and protected.

Cons:

1. Complexity: Tokenizing a writing style is a complex process that requires advanced natural language processing (NLP) techniques. It involves capturing subtle nuances of writing, such as syntax, tone, and vocabulary, which can be challenging.

2. Evolution of Style: An author's style can evolve over time. A tokenization model created based on earlier works may not accurately represent current or future works, potentially leading to disputes over what constitutes protected content.

Legal Soundness

Pros:

1. Innovative Protection: This approach represents a novel way to protect creative works, potentially extending the scope of copyright law to cover not just the content but also the style of writing.

2. Enforcement: If legally recognized, tokenized styles could provide a concrete basis for enforcement against unauthorized use of an author’s style, adding an extra layer of protection for creators.

Cons:

1. Legal Precedents: Copyright law traditionally protects the expression of ideas, not the ideas themselves or the style in which they are expressed. Establishing legal precedents for style protection via tokenization could be challenging and require significant legal reforms.

2. Ambiguity: There may be ambiguity in defining what constitutes a protectable style and how distinct it must be from others to qualify for protection. This could lead to legal uncertainties and disputes.

Improved, Additional, or Alternative Means

1. Watermarking:

• Technological: Embedding digital watermarks in text can help identify the original author and track unauthorized use. This can be done subtly so as not to alter the content visibly.

• Legal: Digital watermarks can serve as evidence in legal disputes over authorship and unauthorized use.

2. Blockchain:

• Technological: Utilizing blockchain for copyright registration can provide a tamper-proof record of when and by whom a work was created. This can be particularly effective for establishing ownership and the originality of works.

• Legal: Blockchain records can be used in court to prove the provenance of a work, making it easier to enforce intellectual property rights.

3. Licensing Agreements:

• Legal: Authors can use licensing agreements to explicitly define the permissible use of their writing style. These agreements can include clauses that prevent the use of their style without permission, thereby protecting their intellectual property.

4. Enhanced Copyright Frameworks:

• Legal: Advocating for reforms in copyright law to specifically address and protect creative styles can provide a more straightforward legal basis for protection. This could involve creating new categories or extending existing ones to cover stylistic elements.

5. AI Monitoring Tools:

• Technological: Developing AI tools that monitor the internet for unauthorized use of a particular writing style. These tools can alert authors to potential infringements, allowing them to take action promptly.

Conclusion

The idea of using tokenization to copyright a writing style is a forward-thinking approach that leverages technological advancements to address emerging challenges in intellectual property protection. While it presents potential benefits, there are significant technological and legal hurdles to overcome. Combining this approach with other methods like watermarking, blockchain, licensing agreements, and enhanced copyright frameworks may offer a more robust and comprehensive solution for protecting authors' creative works in the age of AI.

Claude’s answer:
The news item you've described raises some interesting points about protecting creative works in the age of AI, but also presents several challenges and potential issues. Let me break down my thoughts:

1. Technological Feasibility: It's unclear from the description how exactly Dr. Morrison has "tokenized" his writing style in a way that can be copyrighted. Writing styles can be quite subjective and nuanced, and it's not immediately obvious how one could reduce a style to a set of tokens or subwords in a way that comprehensively captures its essence.

2. Legal Considerations: Copyrighting a writing style itself is a novel concept, and it's uncertain whether current copyright laws would recognize and protect such a claim. Copyrights traditionally apply to specific creative works (books, articles, etc.), not to abstract styles or techniques.

3. Enforcement Challenges: Even if such a copyright were granted, enforcing it could be extremely difficult. How would one definitively prove that an AI system has replicated the copyrighted "style" in a way that infringes on the copyright? This could lead to a lot of legal ambiguity and potential disputes.

4. Impact on AI Development: If such stylistic copyrights became widespread, it could potentially hinder the development of AI systems that aim to emulate or learn from existing writing styles, which is a key aspect of how many language models operate.

Potential Alternatives or Additional Measures:

1. Relying on Traditional Copyrights: Authors can continue to rely on traditional copyrights for their specific creative works (books, articles, etc.). While AI systems may be able to emulate styles, they cannot directly copy copyrighted content without permission.

2. Watermarking or Fingerprinting: Techniques like digital watermarking or content fingerprinting could be used to embed unique identifiers or signatures into creative works, making it easier to detect and prove instances of direct copying or plagiarism by AI systems.

3. Ethical AI Development: Encouraging and enforcing ethical practices in the development of AI language models, such as requiring explicit permission from authors before training on their copyrighted works, could help balance the interests of creators and AI developers.

4. Regulated Access to Training Data: Implementing regulatory frameworks that govern the use of copyrighted materials for AI training purposes, potentially through licensing or compensation mechanisms, could provide a middle ground between protecting creators and enabling AI development.

5. Public Domain and Fair Use: Promoting the use of public domain and fair use materials for AI training could help mitigate some of the concerns around copyright infringement, while still allowing AI systems to learn from a diverse range of writing styles.

Ultimately, while Dr. Morrison's approach is novel, there are significant technological and legal uncertainties surrounding its viability and potential implications. A balanced approach that considers the interests of creators, AI developers, and the broader public interest may be necessary to navigate this evolving landscape effectively.

Google Gemini’s answer:
Can Copyrighting a Writing Style Work?

There are reasons to be skeptical of Dr. Morrison's approach:

• Copyright Protects Expression, Not Style: Copyright protects the specific way ideas are expressed, not the general style. It's hard to copyright a broad concept like "writing style." Common phrases, vocabulary choices, or sentence structures wouldn't qualify.

• Copyright Office Scrutiny: The Copyright Office might scrutinize such a claim. They might argue the tokenization process is an idea, not expression, and ideas aren't copyrightable.

Pros and Cons of This Approach

Cons:

• Limited Protection: Even if granted, the copyright might offer narrow protection. Someone could write in a similar style without infringing.

• Enforcement Challenges: Proving someone copied your style could be very difficult and expensive in court.

Pros:

• Potential Benefits: If upheld, it could offer some protection and give authors a sense of security.

Improved Approaches for Authors

Here are some more reliable ways for authors to protect their work:

• Copyright Registration: Registering your copyright with the U.S. Copyright Office provides a stronger legal position in case of infringement.

• Digital Rights Management (DRM): This technology can control how your work is accessed and used online.

• Attribution and Licensing: Clearly mark your work with a copyright notice and consider using Creative Commons licenses to specify how others can use it.

• Focus on Originality: The more original your ideas and plot are, the harder it is to copy them.

Conclusion

Copyrighting a writing style is an unproven approach with limited legal weight. Traditional copyright registration,combined with proactive measures like attribution and licensing, offer a more reliable path for authors to protect their work.

MS Copilot’s answer:
The concept of copyrighting a specific style of writing, particularly in the context of AI and tokenization, is indeed a novel approach. While I don’t have personal feelings, I can provide an analysis based on current knowledge and practices in the field.

Technological Perspective: From a technological standpoint, tokenization is a process where text is broken down into smaller units for processing. AI can use these tokens to learn and generate text in a particular style. Copyrighting a tokenization method could potentially protect the unique style encoded within these tokens, making it difficult for others to replicate the style without permission.

Legal Perspective: Legally, the situation is more complex. Copyright law traditionally protects the expression of ideas rather than the ideas themselves or the methods used to create them. However, if a tokenization method is so closely tied to a specific style that it becomes a unique form of expression, there might be a case for copyright protection. This would be a significant shift in copyright law, recognizing the process as a protectable part of the creative work.

Pros:

• Protection of Creative Expression: Authors could safeguard their unique styles from being replicated without consent.

• Monetization: It could open new avenues for creators to monetize their styles or methods.

Cons:

• Innovation Stifling: Overly broad copyrights could hinder creativity and innovation in AI development.

• Enforcement Challenges: It may be difficult to enforce such copyrights, especially when dealing with AI that can learn from vast amounts of data and potentially mimic styles inadvertently.

Caveats:

• Legal Precedent: There is little legal precedent for such copyrights, and laws may need to evolve to accommodate these new concepts.

• Technological Advancement: As AI technology advances, it may become increasingly challenging to attribute a style to a single creator or method.

Alternative Means:

• Trade Secrets: Keeping the tokenization method confidential as a trade secret might be a more practical way to protect it.

• Patents: If the method is novel and non-obvious, seeking a patent might provide legal protection for the process itself.

• Licensing Agreements: Creators could use licensing agreements to control how their styles and methods are used by others.

In conclusion, while the approach of copyrighting a writing style via tokenization is intriguing, it raises several questions and challenges that would need to be addressed both technologically and legally. It’s a developing area where the intersection of AI technology and intellectual property law will likely continue to evolve¹²³⁴.