Artificial Intelligence and Copyright law in India: Challenges, Judicial Development, and the Need for Reform

Introduction

Artificial Intelligence (AI) plays a crucial role in human life. It helps make people’s lives easier, especially Generative AI, which can create text, images, audio, and videos.

In India, the protection of creative works is primarily governed by the Copyright Act, 1957[1], which governs copyright protection for eligible works and identifies authors under its provisions. The objective of the Act is to protect the original expression of an idea rather than the idea itself.

However, one important question remains unanswered: who owns an AI-generated work? Is it the developer, the user, or no one? If no one owns such a work, does it become part of the public domain? Another important question is whether AI-generated works are eligible for copyright protection under Indian law.

This article examines how the provisions of the Indian Copyright Act deal with the protection of creative works in the context of Generative AI. It also highlights the gaps in the existing legal framework and briefly discusses relevant international developments.

Understanding Artificial Intelligence

Artificial Intelligence (AI) is a type of technology that performs tasks based on a user’s instructions. It can generate text, images, audio, videos, and other forms of content according to the prompts provided by the user. In recent years, there has been a significant rise in the use of AI tools such as ChatGPT, Gemini, Perplexity, and many others.

AI can create new content or improve existing content based on the user’s instructions. Earlier, AI was mainly used for limited and specific tasks. However, with technological advancements, Generative AI has emerged as a more advanced form of AI.

Generative AI uses different types of input to generate a single output that resembles content created through human intelligence. It helps speed up various processes by automating creative tasks, increasing productivity, generating ideas and drafts, and making advanced technology more accessible to people.

Many people believe that AI is a recent development. However, its history goes back much further. In 1837, Charles Babbage designed the Analytical Engine, the first general-purpose mechanical computer, with the support of his colleague Ada Lovelace, who is widely recognized as the world’s first computer programmer.

In 1950, Alan M. Turing published his landmark paper Computing Machinery and Intelligence[2], in which he introduced the Turing Test to evaluate machine intelligence and raised the famous question, “Can machines think?”

Copyright Law under the Copyright Act, 1957

The Copyright Act, 1957 has been amended six times, with the most recent amendment made in 2012. The Act provides copyright protection for literary works, dramatic works, artistic works, cinematograph films, and sound recordings. It grants legal protection to copyright owners, encourages creativity and innovation, and safeguards original works from unauthorized copying and infringement.

The Copyright Act also recognizes the moral rights of authors. Under section 2(d)[3], the Act defines who is considered the author of different categories of works. Further, under section 17 of the Copyright Act[4], the author is generally regarded as the first owner of the copyright, subject to the exceptions provided under the Act. The concept of authorship under the Act is based on human creativity, skill, labor, and intellectual effort.

Whether AI-Generated Works Can Receive Copyright Protection

Under the current legal position in India purely AI- Generated work generally do not receive copyright protection because the law recognizes human authorship. In India, AI-generated works do not receive copyright protection because copyright requires a human element in the creation of a work. AI is generally considered a tool or a computer program that generates outputs based on the data on which it has been trained and the prompts provided by the user. Therefore, a work created entirely by AI lacks the human authorship required for copyright protection.

Similarly, in the United States, human authorship is also required for copyright protection[5]. However, the issue of copyright in AI-generated works remains a subject of debate in many jurisdictions, including the United States, the United Kingdom, and the European Union. In the case of AI-generated content, the user generally provides only a prompt, while the AI generates the output after analyzing a large amount of data. Therefore, merely providing a prompt is generally not considered sufficient human contribution to claim copyright in a purely AI-generated work.

Who Should Be Considered the Owner of AI-Generated Content: The User, the Developer, or No One?

At present, this question remains unanswered by the Indian legislature. There are no clear legal guidelines regarding the authorship or ownership of AI-generated works. Since copyright law requires human authorship, the ownership of purely AI-generated content remains uncertain under the current legal framework.

However, a user may able to claim copyright in an AI-assisted work where the user has made significant human contributions, such as designing, modifying, organizing, or editing the final output. The purpose of copyright protection is to recognize and reward human creativity and intellectual effort.

The Copyright Act, 1957 also contains provisions relating to literary works. Under section 2(o)[6], the Act defines a “literary work” to include computer programs, tables, compilations, and computer databases. This shows that the scope of literary works is not limited to poems or novels but also includes other original written works, such as books, articles, and research papers, provided they satisfy the requirements of copyright law.

Whether training AI models on copyrighted works amounts to copyright infringement

Developer train the AI models from the thousands of copyrighted documents or works which helps to provide the output to the user this is a debatable issue because in Indian copyright act does not provide a clear provision about this

In USA copyright infringement is based on the doctrine of ”Fair use” [7]in this it provide a limited utilization of copyrighted material without the permission of copyright owner for some societal benefits such as Criticism, research, teaching , comments etc. in all of this there aim is to provide the awareness about the work without infringing the right of an owner but before that we have to take care of 4 factors firstly the purpose and character of the use of the work whether is it commercial or educational etc, secondly the nature of the copyrighted work, thirdly the amount of copyrighted work taken and  substantiality used or not any lastly what is the impact on the market of this work. This doctrine expand the knowledge and idea of copyrighted work for social welfare purpose with protecting the right of author.

In  India  the answer on training AI models on copyrighted works amounts to copyright infringement is a vague its answer is not properly answered its neither black or white more we can say it’s a grey area where  there is no such guidelines and provisions regarding this.

But as per the government’s department for promotion of industry and internal trade (DPIIT) issued a working paper [8]that conclude commercial AI training does not qualify for traditional  ”fair dealing.”

There is a exceptions in India regarding the infringement of copyrighted work as per section 52 [9]of copyright act for specific purposes such as individual study, research, teaching  purpose etc. copyrighted data can be used without permission of an author.

International Developments and Important AI Copyright Case

In the case of Getty Images (US) Inc v Stability AI Ltd [2025] EWHC 2863 (Ch)[10], the High Court observed that AI models do not store or reproduce the original copyrighted content for users in its original form. AI-generated output is produced by processing and analyzing a large amount of copyrighted and non-copyrighted material used during training. Therefore, training data is not necessarily reproduced as an infringing copy in the output.

However, AI companies may still be liable for trademark infringement if their systems generate protected trademarks, logos, or watermarks in a manner that infringes the rights of trademark owners.

Gaps in the Indian Legal Framework

One of the major gaps in the Indian legal framework is the lack of recognition of non-human authorship. The Copyright Act, 1957 does not contain any specific provision that determines who should be regarded as the author of an AI-generated work. Under the existing legal framework, copyright protection is generally based on human authorship and human intellectual effort. Therefore, the copyright status of purely AI-generated works remains uncertain. This issue arises because Generative AI creates content by processing and analyzing large amounts of data.

Broadly, there are two major gaps in the current legal framework. The first is the input-side issue, which arises during the training of AI models. Generative AI requires a massive amount of data for training, including both copyrighted and non-copyrighted material. This raises concerns regarding the use of copyrighted works without the permission of copyright owners.

The second is the output-side issue, which relates to authorship. Under section 2(d)[11] of the Copyright Act, authorship is linked to a human author. As a result, when content is generated entirely by AI with little or no human creative contribution, questions arise regarding who should be recognized as the author and whether such works are eligible for copyright protection.

Recommendations for Reforms

A compulsory licensing framework can be used to protect the rights of copyright holders. It can help prevent the unauthorized use of copyrighted works while allowing limited access to such works in exceptional circumstances.

A compulsory licensing framework can also create a balance between the rights of copyright holders and the public interest by permitting the use of copyrighted works under legally recognized conditions.

Conclusion

At present, there is no specific legal provision regarding the authorship of AI-generated works, which creates uncertainty in determining the ownership of such creations. In traditional copyright law, the person who creates the work is recognized as the author and receives copyright protection. However, in the case of AI-generated works, the question of ownership remains unanswered. It is still uncertain whether the developer, the user, or no one should be regarded as the owner of such works.

This article recommends the adoption of a compulsory licensing framework to help address issues relating to the unauthorized use of copyrighted works. Such a framework may help achieve a fair balance between the rights of copyright holders and the interests of the public.

Bibliography

Table of Case

Getty Images (US) Inc v Stability AI Ltd [2025] EWHC 2863 (Ch).

Table of Legislation

Copyright Act 1957.

United States Copyright Act 1976.

Official Publications

Department for Promotion of Industry and Internal Trade, Issues for Copyright and Artificial Intelligence (Working Paper).

US Copyright Office, Copyright Registration Guidance: Works Containing Material Generated by Artificial Intelligence (16 March 2023).

Journal Articles

Turing AM, ‘Computing Machinery and Intelligence’ (1950) 59 Mind 433.

[1] Copyright Act 1957.

[2] Alan M Turing, ‘Computing Machinery and Intelligence’ (1950) 59 Mind 433.

[3] Copyright Act 1957, s 2(d).

[4] Copyright Act 1957, s 17.

[5] US Copyright Office, Copyright Registration Guidance: Works Containing Material Generated by Artificial Intelligence (16 March 2023).

[6] Copyright Act 1957, s 2(o).

[7] 17 USC § 107.

[8] Department for Promotion of Industry and Internal Trade, Issues for Copyright and Artificial Intelligence (Working Paper).

[9] Copyright Act 1957, s 52

[10] Getty Images (US) Inc v Stability AI Ltd [2025] EWHC 2863 (Ch).

[11] Copyright Act 1957, s 2(d)