As AI adoption accelerates, regulators must adapt laws on data training and IP ownership to keep pace with technological advancements.
As Artificial intelligence (AI) usage grows, its share in the global media and entertainment market is expected to rise to $99.48 billion by 2030, registering a CAGR of 24.2% from 2025 to 2030. The AI market size was already $680 million in 2022 and is expected to reach $3.94 billion by 2028, with a CAGR of 33.28% between 2023 and 2028. There are concerns regarding data privacy and securing intellectual property rights of the original creators that AI draws its training from. However, AI also lowers costs while also expanding revenue streams and helping to reach a broader audience and increase the accessibility of content.
The extant IP law in India vis-a-vis AI
At present, the Indian laws on intellectual property indicate that only legal or natural persons can have ownership over IP created with the use of AI. The Copyright Act grants exclusive rights to authors of literary, musical, artistic and cinematic works to reproduce, distribute and monetize their creation. It also ascribes ownership to the author of the computer program, where any such work is computer generated i.e., the legal person who created the computer program .
Generative AI trains itself by using the data provided by users, including data which may otherwise be protected under the Copyright Act. This raises 2 vital questions – (i) is the use of data for training AI protected by the fair dealing provisions; and (ii) if yes, how should this affect the ownership of such works as stipulated under the law. The answers to both these questions have become the subject matter of litigation globally.
OpenAI, the company that created ChatGPT, is facing lawsuits in New York for training their AI on copyrighted material without requisite permissions. OpenAI has in turn denied allegations of accessing subscription-based material and instead stated that its Large Language Model (LLM) is only trained on publicly available data.
In India, ANI News has filed a similar lawsuit against OpenAI for using its published content to train the AI model and have further alleged that several news articles were verbatim reproductions of ANI news content. The court must determine whether using content to train LLMs constitutes infringement or qualifies as fair dealing under Indian law. Notably, the Union Minister of State for Commerce and Industry, Shri Som Parkash, in an unstarred question in the Rajya Sabha, has clarified that unless Generative AI is being used for fair dealing purposes, authorization will be needed before training LLMs with copyrighted content.
In 2020, artist Ankit Sahni used an AI tool, RAGHAV (Robust Artificially Intelligent Graphics and Art Visualizer), to create an artwork titled Suryast and applied for its copyright registration. The application was initially rejected due to the absence of a human author. However, it was later accepted after Ankit was added as a co-author. Subsequently, a notice of withdrawal was issued, seeking clarification on RAGHAV’s legal status, ultimately reinforcing the requirement that authorship be attributed only to human creators.
As of now, The Trade Marks Act, 1999 does not contain any provisions for the recognition of a trademark generated by AI, and currently only recognized natural and juristic persons as owners of a trademark. Taking the Patent Act, 1970 in mind, AI inventions fall within the purview of Section 3(k), which excludes mathematical methods, business methods and computer programs or algorithms from protection. However, in 2019, the Delhi High Court held that a patent would not lose its protection under the Patent Act, 1970 simply because it is computer generated. It was re-affirmed in 2024 that a computer-based program or invention may be patentable if it provides a technical effect or contribution such as solving a technical problem or enhancing a technical process. Therefore, AI-generated works may be eligible for registration, provided they demonstrate technical utility as outlined above. However, ownership will likely be attributed to the legal entity responsible for the AI’s creation.
AI and the Indian regulatory landscape
The National Strategy for Artificial Intelligence report released by NITI Aayog in 2018 has recognized the need for a robust IP framework that fosters AI innovation. The report recommended for the Ministry of Commerce and Industry, and the Department for Promotion of Industry and Internal Trade to set up a task force to modify the existing IP regime to support the growth of AI in the Indian tech-ecosystem. It also suggested that the task force establish guidelines to ensure that the use of AI is ethical, while promoting data privacy, transparency and accountability of the AI systems being used.
Further, the Ministry of Electronics and Information Technology’s Subcommittee Report on AI Governance Guidelines states that the access to larger data sets for the purpose of training AI systems are among the 5 reasons that have led to the exponential growth of AI. In light of the same, the sub-committee report recognised the need to legislate on whether AI systems should be allowed to train on bulk datasets that may include copyrighted data, without taking approval from each copyright holder and the requirement of human authorship for copyright protection.
As it stands, the sub-committee’s report along with NITI Aayog’s National Strategy for Artificial Intelligence, Principles for Responsible AI and Operationalizing Principle for Responsible AI, are India’s primary guidelines for the use of AI.
Global trends and the road ahead for India
The United Nations Conference on Trade and Development in their report, ‘Digitalisation, Artificial Intelligence and the Creative Economy’, have underscored the significance of AI in producing, distributing and consuming content in the creative economy. They recognized the growth of AI in creative industries like art, the film industry and video games, resulting in the increased contribution of AI towards the creative economy. The creative economy is a $30 billion industry in India and accounts for employing nearly 8% of India’s working population.
Union Minister Ashwini Vaishnaw has emphasized that advancements in AI technologies will bring significant disruption to the creative industry, while reiterating that AI’s impact on intellectual property presents a challenge for lawmakers. He has stressed the importance of safeguarding the rights of original creators to ensure continued innovation and incentivization.
Legislators the world over have recognized the need to govern IP produced by AI and are following different approaches. A case in point is the UK Government who has published a consultation paper — Copyright and Artificial Intelligence, which examines proposals to change UK’s copyright protection framework. Presently, there is an exception on copyright protection for text and data mining when copyrighted content is being used only for non-commercial purposes. Japan allows the use of datasets for commercial purposes as well as training AI models. Similarly, Singapore has also allowed the use of copyrighted content without permission from the authorized owner for computational data analysis. It also extends the usage for commercial purposes to train AI if the user has lawful access. These developments indicate a global trend to develop an AI-based economy on the principle of transparency.
The need of the hour therefore for Indian regulators lays in striking a critical balance between driving innovation and ensuring equitable access to technology, fostering both progress and inclusivity in the digital age.
