The AI IP Conundrum: Ownership & Inventorship

Abstract:

The artificial intelligence (AI) culture has posed unprecedented challenges and hurdles to the traditional framework of intellectual property (IP) laws, particularly in the area of inventorship. An era of completely new culture, new issues, new judgements is yet to come forward.

Introduction:

The concept of IP protection is quite dynamic and rapidly evolving over centuries to encourage innovation and creativity. Traditionally, inventorship was restricted to human beings, who have been recognized as the originators of intellectual works. However, the evolution of AI has introduced a new face to the picture, where machines can generate creative and produce required outputs. The involvement of AI is necessary as well as unavoidable in our day-to-day work. Even for the smallest of tasks, even on social medias, every other mobile application we get access to an AI-BOT. and when this comes to areas of inventorship or creativity or when the artists are using AI, it arises many significant question marks. This raises fundamental questions about the nature of inventorship and the applicability of existing IP laws to AI-generated creations.

The most recent significant update to Indian patent law was the Patent Amendment Act, 2021. This Act introduced several amendments to the Patents Act, 1970, with the primary aim of strengthening IP protection and promoting innovation. The basic frame work of IP laws was based on the Patents Act, 1970, and the Copyright Act, 1957, for the last few decades. While these laws were designed to protect human-created inventions and works, their applicability to AI-generated content was a subject of question. The absence of specific provisions addressing AI-generated inventions created a legal vacuum, leading to uncertainty and potential disputes.

There were several questions raised and borne out of the evolution of AI that can only be answered with help of planned precedents, new laws and the interpretation by SC, including:

• Establishing a clear definition of inventorship in the context of AI is crucial. Should it be limited to human beings, or can AI systems be considered inventors?
• Determining ownership rights to AI-generated inventions is another complex issue. Should the rights vest with the AI developer, the owner of the AI system, or the AI itself?
• How to encourage AI research and development without compromising the principles of IP protection?

Case laws:

Case 1: Li v Liu[1] (2023)

• Copyright of AI-generated content: The court ruled that AI-generated content can be copyrighted if it reflects the author's personalized expression.
• Author's role: The author must have selected, depicted, and set parameters for the content.
• AI's role: AI is a tool, not a copyright holder.
• Copyright holder: The user who provided intention, ideas, and intellectual input is the copyright holder.

Case 2: Copyright infringement of Ultraman’s image in 2024[2]

• AI company liability: The AI company infringed copyright by generating images substantially similar to copyrighted works.
• Duty of care: AI service providers must establish a complaint mechanism, warn users of risks, and label AI-generated content.
• Damages: Damages awarded consider the seriousness of infringement, expenses, and the need to balance copyright protection with AI industry development.

In OpenAI vs. Scarlett Johansson[3]

OpenAI's voice assistant sounds like Scarlett Johansson. Kristelia García, a professor at Georgetown Law, analyses the potential legal arguments for and against OpenAI's actions.

García explained the key legal issue and violation of Johansson's right of publicity. She discusses the concept of rights of publicity and how it applies to celebrities like Johansson. García also references the Midler v. Ford case, where a similar situation occurred.

There must be a line between illegal use of someone's voice and the creation of a voice that sounds like them. García discusses the ethical concerns and the potential need for regulations regarding AI-generated content. She also analyses the implications of media companies giving OpenAI access to their content and the potential disruption to the media industry if more people rely on AI chatbots for news.

Suggestions

• International Collaboration: Given the global nature of AI development, international cooperation is essential to establish consistent and harmonized legal frameworks.
• Education and Awareness: Raising awareness among AI developers, businesses, and policymakers about the IP implications of AI is crucial.
• Case Studies: Analysing real-world cases involving AI and IP can help to identify emerging trends and challenges.
• Ethical Considerations: Incorporating ethical principles into AI development and IP protection can help to address concerns about bias, fairness, and accountability.

Problems

• Traditionally, both developers and AI models could be liable for IP infringement.
• Current IP laws (Copyright Act, Trademark Act, etc.) struggle to address AI due to its self-learning nature.
• AI (especially Generative AI) raises questions about who owns the content it creates.
• Indian courts haven't recognized AI models as authors (similar to the US).
• A recent lawsuit highlights the legal grey area regarding AI-generated content.
• India needs to update IP laws to consider AI's role in content creation.

Main Argument:

• Existing IP laws are outdated for AI technology.
• Unclear legal definitions create a risk for both creators and users of AI-generated content.
• India needs to adapt its legal framework to address these issues.

Acts and Laws

While India doesn't have specific laws directly addressing AI and IP, several existing laws may be relevant:

• Patents Act, 1970[4]: This act governs the grant and protection of patents. While primarily focused on human-created inventions, it may need to be adapted to accommodate AI-generated works.
• Copyright Act, 1957: This act protects original literary, artistic, and dramatic works. It may be applicable to certain AI-generated content, such as software code or artistic expressions.
• Trade Secrets Act, 1999: This act protects confidential information that has commercial value. It may be relevant to AI algorithms or data that are kept secret.

Author’s POV:

Technological advancements have blurred the lines between the real and digital worlds. AI and software are increasingly involved in creating innovative products, leading to questions about ownership and intellectual property rights.

Current Indian laws struggle to recognize AI as an author or inventor. While the Copyright Act partially addresses computer-generated works, it doesn't explicitly recognize AI's authorship. The Patent and Design Acts also lack provisions for AI-related innovations.

International laws offer some guidance. Countries like the UK and New Zealand have provisions for copyright protection of computer-generated works, even without a human author.

The Turing Test highlights AI's capabilities. Alan Turing's test evaluates a machine's ability to mimic human intelligence. While AI has evolved significantly, questions about IP rights for AI-generated works remain unresolved.

India's Parliamentary Standing Committee recommends changes. The committee suggests introducing a separate category for AI-based inventions and reviewing existing laws to accommodate AI and its creations.

Balancing regulation and innovation are crucial. The legal framework needs to adapt to AI's capabilities while protecting the rights of creators, both human and machine.

Conclusion

The intersection of AI and intellectual property is a complex and rapidly evolving field. As AI continues to advance, it is essential to develop legal frameworks that can effectively protect innovation while addressing the unique challenges posed by AI-generated creations. By fostering international collaboration, promoting education and awareness, and addressing ethical concerns, policymakers can work towards creating a legal environment that supports the development and use of AI while ensuring the protection of intellectual property rights.

Autor: Sneha 3RD year Law Student, Department of Law, Kurukshetra University.

[1] China’s First Case on Copyrightability of AI-Generated Picture. (2023, December 7). KWM. Retrieved September 30, 2024, from https://www.kwm.com/cn/en/insights/latest-thinking/china-s-first-case-on-copyrightability-of-ai-generated-picture.html

[2] AI generated content platform found to infringe Ultraman copyright by internet court in first judgment of its kind. (2024, March 7). Lexology. Retrieved September 30, 2024, from https://www.lexology.com/library/detail.aspx?g=5746ee79-1576-4dc4-9ce9-0aa2212a3045#:~:text=Shanghai%20Xinchuanghua%20Culture%20Development%20Co,a%20lawsuit%20against%20the%20platform.&text=In%20February%202024%2C%20Guangzhou%20Internet,the%20grounds%20of%20copyright%20infringement.

[3] OpenAI v. Scarlett Johansson? Georgetown law professor answers legal questions on AI-Generated content. (2024, June 4). Georgetown.edu. Retrieved September 30, 2024, from https://www.georgetown.edu/news/ask-a-professor-openai-v-scarlett-johansson/

[4] Patents Act 1970. (2015, March 11). india.gov.in. Retrieved September 30, 2024, from https://ipindia.gov.in/writereaddata/Portal/IPOAct/1_31_1_patent-act-1970-11march2015.pdf

Image- Understanding the distinction between inventorship and ownership in patent law. (2024, July 24). LinkedIn. Retrieved September 30, 2024, from https://www.linkedin.com/pulse/understanding-distinction-between-inventorship-ownership-reinier-smit-r4ube/