Generative Artificial Intelligence And Copyright Laws In India: The Nuances Of Authorship And Fair Dealing
Gaurav Mongia, School of Law, GD Goenka University, Gurugram, Haryana, India
ABSTRACT:
The advancement of information technology in the form Large Language Models (LLMs) or Artificial Intelligence models, also known as Generative AI, has given rise to important legal questions concerning Authorship, Ownership, Originality and the usage of copyrighted work for the training of AI Models. The Indian Copyright Act of 1957 is based on the traditional approach that the creative work originated from human intellectual efforts. However, the AI models such as Open AI’s CHATGPT, Google’s Gemini etc. can now produce literary, musical, artistic and other forms of content with very limited involvement of human. This development creates an uncertainty with respect to the recognition and ownership of AI-Generated works under the Indian Copyright Act, 1957. The primary statutory framework for considering these issues in India is provided by Sections 2(d), 13, 14, 17, 51 and 52 of the Copyright Act. These provision raises questions about the role of AI developers, users and other human participants. Likewise, the use of copyrighted works to train AI models raises questions of reproduction, infringement and the scope of statutory exceptions.
This research paper discusses if the use of copyright materials for AI training constitutes actionable infringement under Section 14 or is fair dealing under Section 52. It also questions whether the law created for human inventors can accept AI or the humans writing text prompts as "authors" from the context of Section 2(d) of the Act. Finally, the paper examines recent court decisions (Specially, the Delhi High Court decision on Interlocutory application in ANI Media Pvt Ltd v. OpenAI) to see how judges are ruling on these disputes as they are occurring.
Keywords: Generative Artificial Intelligence, Authorship, Ownership, Fair dealing, Copyright.
