Copyright Protection and AI Authorship: Navigating Indian Law for AI-Generated Works

AI and Copyright Law in India: Key Insights

In the rapidly evolving world of artificial intelligence (AI), questions about creativity, ownership, and protection are more pressing than ever. As AI tools generate art, music, code, and text at unprecedented speeds, the intersection of AI and copyright law in India demands careful examination. Can AI be an author? Does training AI on copyrighted material constitute infringement? This blog post delves into these issues, drawing from established Indian copyright principles and emerging challenges.

Disclaimer: This article provides general information and is not intended as legal advice. Consult a qualified attorney for specific guidance.

Overview of Copyright Law in India

Copyright law in India, governed primarily by the Copyright Act, 1957, protects the rights of authors and creators against unlawful reproduction or exploitation of their works. It encourages creativity by offering a long period of protection for authors and artists.

Pee Pee Publishers & Distributors (P) Ltd. vs Neena Khanna - Delhi (2010)

A key requirement is originality. Under Section 2(o) of the Act, a computer program qualifies as an original literary work, making it eligible for copyright protection. Infringement arises if someone alters or reproduces another's computer program without permission. 2005 0 Supreme(AP) 683

Section 14 defines copyright as the exclusive right to perform certain acts on a 'work', while Section 16 clarifies that no copyright exists outside the Act's provisions. 2014 0 Supreme(Del) 1387

The philosophy underpinning this is profound: The law does not permit one to appropriate to himself what has been produced by the labour, skill and capital of another. This reflects Article 27 of the Universal Declaration of Human Rights and forms the foundation of copyright law. 2010 0 Supreme(Bom) 116

Key Legal Principles: Idea-Expression Dichotomy and Infringement

A cornerstone of Indian copyright law is the idea/expression dichotomy. Copyright does not protect ideas, themes, or facts—only their expression. This principle was firmly established in the Supreme Court landmark case R.G. Anand v. M/s. Delux Films. 2018 0 Supreme(Manipur) 22

The court ruled that similarities must be substantial and material to constitute infringement. If a new work shares a theme but expresses it differently, it typically does not infringe. A literal imitation is required, and material dissimilarities can negate claims, as coincidences may be incidental. 2018 0 Supreme(Manipur) 22

This aligns with global axioms: The most fundamental axiom of copyright law is that 'no author may copyright his ideas or the facts he narrates.' 2013 0 Supreme(Del) 1078

Ownership vests first with the author, independently of registration. 1985 0 Supreme(AP) 247 Registration aids enforcement but is not mandatory; rights can be proven through other evidence.

Dattatray Bapu Dighe VS State of Maharashtra - Bombay

Infringement under Section 63 criminalizes unauthorized reproduction, requiring proof of copying or unauthorized use. Courts assess ownership, assignments, and rights scope rigorously. 2023 0 Supreme(Kar) 494 2023 0 Supreme(Guj) 1305

The Intersection of AI and Copyright Law

AI-generated works challenge traditional notions. Does output from tools like DALL-E or ChatGPT qualify for protection? Indian law emphasizes human authorship and originality tied to expression.

Authorship and Ownership: Copyright typically requires a human creator. AI lacks legal personhood, so rights may vest in the developer, user, or remain unprotected unless human input is substantial.
Training Data and Infringement: AI trained on copyrighted datasets risks infringement if outputs substantially imitate originals. However, under the idea-expression rule, transformative new expressions may avoid liability. 2018 0 Supreme(Manipur) 22

For instance, copyright in films does not automatically extend to underlying works; it must vest per the Act. 2023 0 Supreme(Del) 4240 Co-existing copyrights are familiar, as in performances under limited licenses. 2011 0 Supreme(Bom) 857

Digital works, including multimedia, enjoy specific protections. Section 13(4) safeguards contributors like musicians, highlighting equitable recognition. 2024 0 Supreme(Cal) 1378

Boundaries with design laws further clarify: Works registered under the Designs Act lose copyright protection to prevent overlap. 2025 0 Supreme(SC) 628 2025 0 Supreme(Mad) 4647 2024 0 Supreme(Guj) 2046

Challenges Posed by AI-Generated Content

AI complicates infringement assessments. Outputs may resemble training data, but dissimilarities could defend against claims. Courts might evaluate if AI merely regurgitates or creates novel expressions.

Copyright cannot be conferred by consent outside the law; it must align with statutory criteria. 2023 0 Supreme(Del) 4240 2014 0 Supreme(Del) 1387

As AI blurs lines, questions arise: Is AI output 'original'? Legal frameworks may need evolution, similar to computer programs' recognition as literary works. 2005 0 Supreme(AP) 683

Practical Implications and Enforcement

For Creators and Businesses: When using AI, document human contributions to claim authorship. License training data where possible.
Litigation Trends: Courts dismiss improper claims, emphasizing procedural compliance. 2023 0 Supreme(Guj) 1305
Global Context: Insights from texts like World Copyright Law underscore harmonization needs. 2008 4 Supreme 98

Legal procedures favor evidence-based enforcement, even without registration.

Dattatray Bapu Dighe VS State of Maharashtra - Bombay

Conclusion and Key Takeaways

The intersection of AI and copyright law in India hinges on protecting expressions, not ideas. While established principles like those in R.G. Anand provide guidance, AI demands clearer rules on authorship and infringement. 2018 0 Supreme(Manipur) 22

Key Findings:- Copyright safeguards original expressions, requiring substantial similarity for infringement. 2018 0 Supreme(Manipur) 22- Human-centric authorship prevails, challenging pure AI outputs.- Dissimilarities and transformations often defeat claims.

Recommendations:- Monitor cases and reforms for AI-specific guidelines.- Advise clients on risks when leveraging AI tools.- Engage in policy discussions to adapt laws.

As AI advances, staying informed is crucial. Indian law offers a robust foundation, but evolution is essential for this dynamic landscape.

References:-

Pee Pee Publishers & Distributors (P) Ltd. vs Neena Khanna - Delhi (2010)