On the news of the Indian Music Industry and music labels seeking to be impleaded as parties in the ongoing ANI v. OpenAI case, Anjali Tripathi writes on the issues raised by the trade body and labels and the overall impact this case can have on the music industry. Anjali Tripathi is a fourth-year law student at JGLS with an interest in IP rights, access to education, and the creative arts.
After ANI, Bollywood Music Labels Have a Face-off with OpenAI
By Anjali Tripathi
As we already know, organisations around the world are now taking the initiative to sue Gen AI models, alleging copyright infringement for using their material to train their models. We previously saw how ANI had sued OpenAI for copyright infringement (See Bharathwaj’s detailed post on this case here!), and now, there is a new development. This has been brought about by some of India’s leading Bollywood music labels, such as T-Series, Saregama India, and Sony Music and the trade body Indian Music Industry (IMI), representing major domestic and international labels. They seek to join the ongoing lawsuit initiated by ANI against OpenAI based on allegations that their copyrighted sound recordings have been used to train OpenAI’s models without authorisation.
The labels contend that the alleged misuse of sound recordings by Open AI poses a serious threat to the music industry in India and the entertainment sector as a whole. With this addition, this case seems to be turning into a defining moment for the boundaries that companies like OpenAI can blur regarding copyrighted material in India. In this post, we will be looking at the key issues that could come up due to this new development and try to understand the trajectory and impact of this ruling.
Key Legal Issues
Under Sections 13 and 14 of the Copyright Act, rights for sound recordings, including adaptation, making a copy, and distribution to the public, are provided. This is distinct from the protection/rights associated with written content or images, which come u/s 13(1)(a).
The arguments of the music labels mainly contend that OpenAI extracts music compositions, sound recordings and lyrics to train their Gen AI models without permission, as reported in the Tribune. The argument about composition seems to be unclear since the rights to these works would technically lie before IPRS and not IMI. Regardless, with respect to the other works, the use by OpenAI would seem like a direct infringement of the copyright of labels under Section 14. OpenAI has defended itself against allegations of copyright infrigement by invoking the principle of ‘fair use’. Under Section 52 of the Indian Copyright Act, fair use extends to matters such as research, review or criticism. However, as discussed by Akshat Agrawal and Sneha Jain, the transformative use defence is not available in the statute, and there exists no judicial precedent regarding AI training under Indian law, making this case all the more important. More importantly, similar defences of fair use have not worked for other tech companies like Telegram and seem like a long shot for OpenAI.
Further, OpenAI’s contention about jurisdiction continues to exist. They argue that Indian courts lack the jurisdiction to decide on the case as they have no servers or permanent office in India, and neither does the training take place here. The two amicus curiae appointed have stated in their submissions that the Delhi HC does have jurisdiction since ANI operates out of New Delhi. Based on the Court’s final say, this would be an important factor in deciding the trajectory and result of this case.
Who Has the Locus Standi or Right to Sue?
Apart from the issues mentioned above, another interesting issue here is whether the Bollywood music labels even have the locus standi to sue OpenAI in the ongoing ANI lawsuit. The original case, as filed by news agency ANI in 2024, had alleged that OpenAI’s ChatGPT was using their content without authorisation to train the AI models. Similarly, the music labels want to join the bandwagon. Previously, even the Federation of Indian Publishers and the Digital News Publishers Association (DNPA) had sought to intervene and be made parties to the suit.
The scope of the suit could be broadened to include these issues. However, it remains to be seen what the court will decide since this may require them to frame additional issues for adjudication as Justice Amit Bansal, overseeing the case (pdf), has already said that affected parties must file individual suits as the current suit by ANI cannot continue expanding. The upcoming hearing on 21st February 2025 will clear out a lot of these procedural issues.
Broader Industry Implications
The ruling of this case won’t just decide on the future of AI companies and the data they can use for training but also the business models of the music labels and, arguably more importantly, the livelihoods of artists, composers and producers of India. The IMI even claimed that this may affect the global industry, too, acting as a dangerous precedent for the acceptance of AI systems being trained on copyrighted materials.
If the court rules in favour of the music labels and ANI, it could set a strong precedent requiring AI models to obtain explicit licenses for the usage of copyrighted material. On the other hand, a ruling in favour of OpenAI would create a risky loophole for AI companies to exploit the copyrighted material, impacting the business of the music labels that generally own copyright over these works.
Not Just Local but Global Impact
This kind of a case is not just a dispute in India as organisations globally are trying to hold AI companies accountable for copyright infringement. In November 2024, the German GEMA, which, similar to IMI, represents composers and lyricists, filed a lawsuit against OpenAI for the unlicensed reproduction of song lyrics by ChatGPT. Similarly, in the USA, the New York Times, Chicago Tribune and others have ongoing suits against OpenAI for similar allegations.
This highlights the increasing urgency for nations to actively establish frameworks that address these issues, much like the European Union is trying to. In India, however, the growth is slow, with other problems taking the front stage.
The Hearing on February 21
The next hearing of this case to look out for is scheduled for 21st February 2025. This will likely result in the Court addressing procedural issues acting as a critical juncture in the case, which will then lead to the more important, substantive questions about copyright law.
This case brings high-stakes issues to the front stage as we see a battle between intellectual property rights protection and the growth of technology in the world. Further, as India continues to emerge as a key player in the global AI landscape (as stated by even Sam Altman, the CEO of OpenAI here) with a projection to grow at a compound annual growth rate (CAGR) of 25-35%, reaching $22 billion by 2027, up from its current value of $7-10 billion as reported by this Nasscom-EY report, it will be interesting to see how this case impacts ChatGPT’s user data with India being the second largest market after the USA.
The main brunt of this case will be felt most by the labels and news agencies that stand to lose the most. AI tools use vast datasets, including seemingly copyrighted material, to train. This threatens the profit earned by music labels and other organisations by devaluing their work and reducing business opportunities. Had they been reached out for proper licensing and rights, it would benefit them legally and financially.
It will also be interesting to see how this differs from stream-ripping cases (instances where the audio/ video is converted to a downloadable audio file from a streaming platform). In Sony v. Yt1s.Com, Justice C. Harishankar restrained such websites, facilitating stream ripping of copyright materials from YouTube. Similarly, the usage of publicly available content from YouTube to train Gen AI models may arguably result in stream ripping (you can read a detailed discussion on this issue here!), which has been deemed unlawful in India. One may wonder whether this may be another issue to be added to the can of worms that the current lawsuit already is.
As India cements its position in the global technology landscape, this pivotal case will act as a litmus test for how the Courts navigate these complex issues at the behest of intellectual property rights and the economic well-being of businesses.
The author would like to thank Swaraj, Praharsh, and an anon reader for their inputs on the piece.