Redefining Boundaries: AI, Copyright, and Intellectual Property Rights
Contributor: Priyanka Khandelwal
The pervasive growth of AI is evident across multiple domains, including industries, software development, art, and literature. As AI continues to exert influence over our workflows and outputs, there arises a compelling necessity to reassess current copyright laws. AI’s expanding role is poised to impact everything from the technology we use to the books we read and the art we experience. Yet, the issue of authorship remains complex. It raises fundamental questions about whether AI systems can genuinely exhibit creativity or if they merely replicate the styles of human creators whose data they were trained on. This ambiguity challenges traditional notions of authorship and ownership, which are deeply rooted in human creativity and intent.
The Necessity of AI-generated work
In its review of India’s intellectual property rights (IPR) regime, the Parliamentary Standing Committee, in Report No. 161, defined Artificial Intelligence (AI) as “A discipline of computer science, mainly aims to develop systems as well as mechanisms that perform the tasks which generally requires human intelligence. It refers to the ability of machines and technologies to perform cognitive tasks like thinking, perceiving, learning, problem-solving and decision-making.” In today’s industrial and market dynamics, securing clear ownership of Intellectual Property (IP) is crucial for individuals and entities, often shaping their business transactions. A major point of contention is whether ownership can be granted to non-human entities such as software and algorithms contributing to product development. Currently, India’s IP laws lack explicit provisions to recognize, and grant ownership to software and algorithms used in creating IP eligible for statutory protection.
Provisions in the Copyright Act, 1957
The Copyright Act of 1957 in India protects various types of creative works such as books, music, and art. However, as technology evolves, questions have arisen about who owns the rights to works created by AI systems and their developers.
According to the Indian Copyright Act, 1957, works that are original and expressed in tangible forms are eligible for protection. While the Act does not directly mention AI-generated content, it does state that the person responsible for causing the work to be created is considered the author for computer-generated works under section 2(d)(vi). This rule has implications for AI-generated content, presenting both opportunities and challenges. On the other hand, the Patent Act of 1970 and the Design Act of 2000 do not recognize programmers or developers as inventors or owners of innovations produced by software, AI, or algorithms. As a result, India’s current legal framework, including the Patents Act of 1970 and the Copyright Act of 1957, does not fully address the complexities related to who should be credited as the creator or owner of works independently created or innovated by Artificial Intelligence.
ChatGPT as an Author
In the context of Indian copyright law, whether artificial intelligence (AI) systems like ChatGPT can be recognized as authors depends on the legal definition of ‘authorship.’ Section 17 of the Indian Copyright Act specifies that only natural persons can claim authorship, explicitly excluding entities such as AI systems. This legal framework centres around human creativity and the ability to express original thought.
According to Section 17, authorship rights are exclusively reserved for natural persons, which means AI entities like ChatGPT cannot be acknowledged as authors under current law. Additionally, the copyright registration process, outlined in Form-XIV, mandates disclosing personal details of the claimant, emphasizing the human-centric approach of the legislation.
The Rise of Artificial Intelligence and Copyright Issues
Another significant challenge facing the intellectual property (IP) community is enforcing copyrights when AI systems are granted authorship rights. This raises concerns about how to hold AI systems accountable for potential copyright infringement, given that they lack legal personhood and cannot bear legal responsibilities. This complex scenario adds layers of uncertainty to the enforcement of copyright laws in relation to AI-generated content. Recently, a class-action lawsuit was filed by a trio of artists against popular AI Image Generators like Midjourney, DeviantArt and Stability AI for infringing their copyright.
Popular AI generators such as DALL-E and Midjourney utilize the diffusion model to create images. This involves taking an image and introducing noise to obscure it, with the AI learning to reconstruct the original image by reversing this corruption process. Determining infringement becomes intricate in AI-generated art because it often produces works that are similar to, but not identical with, copyrighted works. Mere similarity is insufficient to establish infringement, as it could stem from coincidence or both works drawing from common sources. To prove copyright infringement in AI artworks under Indian law, three criteria must be satisfied:
1. The existence of a similar work that is protected under copyright.
2. The infringing act encroaches upon the exclusive rights granted by copyright.
3. The act constitutes actual infringement, as defined by legal standards.
AI Image Generators have also become adept at producing artwork “in the style of” a certain artist. However, as per the Supreme Court in R.G. Anand v. M/S Deluxe Films, copyright owner has no locus standi to sue the defendant in such instances as artistic styles are not copyrightable. Only expressions of ideas are copyrightable.
The RAGHAV Case: AI and Copyright Challenges
In 2020, the Indian Copyright Office encountered a significant case involving an AI system known as ‘RAGHAV’ and its bid to obtain copyright registration for an artwork titled ‘Suryast.’ Initially denied due to the absence of a human author, the painting was later granted protection when a natural person was designated as a co-author alongside ‘RAGHAV.’ This decision was later followed by a withdrawal notice aimed at seeking clarity on the legal status of ‘RAGHAV,’ highlighting the uncertainty surrounding whether AI qualifies as an artist under the Copyright Act.
AI Training Data
Accessing, analyzing, and mining large datasets for training AI inevitably raises concerns about potential copyright infringement. In India, Section 14 of the Copyright Act, 1957 grants copyright owners exclusive rights to reproduce, distribute, translate, and adapt their works. Exceptions listed in Section 52 mitigate infringement claims, such as use for private or personal purposes, criticism, or reporting current events, which fall under fair dealing.
In the realm of AI-generated art, the legality of training data is evaluated based on fair use criteria, which include:
1. The purpose and nature of the use,
2. The nature of the copyrighted work,
3. The amount used in relation to the copyrighted work as a whole,
4. The effect on the potential market for or value of the copyrighted work.
Originality and Human Involvement
In copyright law, protection is extended to artistic works deemed ‘original’ by statutory standards. A landmark case illustrating this principle is University of London Press Ltd v. Tutorial Press Ltd, where a publisher compiled question papers from the University of London. The university alleged copyright infringement, arguing that professors’ intellectual effort in creating the papers warranted protection. The publisher countered that the papers lacked originality as they were derived from existing knowledge. However, the court ruled that even derivative works involving significant effort qualify for copyright protection, emphasizing diligence over creativity.
In Navigators Logistics Ltd. v. Kashif Qureshi, a copyright claim on a computer-generated list was denied due to the absence of human intervention. The court underscored the necessity of human involvement in the creative process for copyright eligibility in India.”
Conclusion
As we strive to strike a balance between countering the challenges and enhance innovations through AI-driven technologies, it becomes essential to find a solution for regulation and incentivisation. The framework needs to be evolved in consonance with the intellectual property framework. As we aim to balance addressing challenges and fostering innovation through AI-driven technologies, it is crucial to develop a framework that supports both regulation and incentivization. The legal framework, particularly in the realm of intellectual property, must evolve accordingly.
This evolution requires a thorough reassessment of existing laws to create an environment that promotes innovation while protecting the rights of creators, whether human or AI. The Parliamentary Standing Committee report has suggested introducing a distinct category for protecting AI-based inventions under intellectual property rights (IPRs). Recognizing the importance and potential of advanced technologies like AI and machine learning in boosting India’s economic growth, the Committee has recommended that the Department for Promotion of Industry and Internal Trade review current legislations, such as the Patent Act of 1970 and the Copyright Act of 1957, to include AI and related innovations within their purview. This approach could set a precedent for requiring references to existing copyrighted works used in training AI systems to create new copyrightable works, thus upholding ethical standards and ensuring compliance with copyright laws.”
