Introduction
As generative artificial intelligence (AI) technologies advance, their impact on various industries, particularly the media sector, becomes increasingly significant. In India, the evolving AI landscape has sparked debates over copyright laws and the monetisation of content used to train AI models. News publishers, who invest heavily in creating high-quality journalism, are advocating for legal reforms to ensure fair compensation for the use of their content in AI training. This push is part of a broader effort to address the legal and ethical complexities surrounding AI, copyright, and content monetisation. This article critically examines these issues, focusing on the Indian context while drawing on global perspectives to highlight the challenges and opportunities that lie ahead.
The Push for Fair Compensation: A Response to Generative AI
The rise of generative AI has introduced new dynamics into the relationship between content creators and AI developers. In India, this relationship is increasingly characterised by tension as news publishers demand compensation for the use of their content by AI systems. The Digital News Publishers Association (DNPA), representing major media organisations such as the Times Group, has been at the forefront of these efforts. They have engaged with various government ministries, seeking amendments to the Information Technology (IT) Rules to prevent potential copyright infringements by AI systems. This advocacy underscores a growing concern that AI models, including those developed domestically, such as Krutrim AI and Sarvam AI’s OpenHathi, may be using copyrighted content without proper authorisation.
The DNPA's call for interim changes to the IT Rules reflects an urgent need to address these concerns before the implementation of the Digital India Act. This upcoming legislation aims to regulate AI and replace the outdated IT Act of 2000, a move seen as essential for aligning India’s legal framework with the rapid pace of AI development. However, this legislative overhaul raises critical questions about how effectively it will balance the interests of content creators and AI developers. The demand for fair compensation is not just a financial issue; it is a matter of ensuring that the creators of valuable content are not marginalised in the AI-driven digital economy.
International Pressures and Evolving Legal Frameworks
The rapid evolution of AI technologies is compelling countries around the world to reassess and strengthen their copyright protections. In August 2023, Japanese industry groups, including the Japan Magazine Publishers Association and the Japan Book Publishers Association, advocated for more robust protections. They argued that existing laws, which permit AI to learn from publicly available data, fall short of safeguarding the interests of copyright holders. Similarly, in May 2024, the U.S. News Media Association, representing around 2,000 publishers, urged U.S. authorities to address AI search engines that allegedly misuse media content. These demands underscore a growing international consensus on the need to adapt copyright laws to the realities of AI.
The legal landscape continues to shift rapidly as generative AI technologies advance. A pivotal case in this context is The New York Times vs. OpenAI. On December 27, 2023, The New York Times filed a lawsuit against OpenAI, accusing it of using its content without authorisation to train its AI models. The newspaper seeks "fair value" for the use of its material, arguing that the compensation provided by OpenAI has been insufficient. OpenAI, on the other hand, contends that its use of data, including articles from The Times, falls under the "fair use" doctrine as defined by U.S. copyright law. This doctrine allows for the reuse of copyrighted material for certain purposes such as research and teaching. OpenAI claims that its use of this content for training AI models is "transformative" and thus qualifies as fair use. However, The Times argues that OpenAI’s use does not meet the criteria for transformative use, suggesting that AI-generated products directly compete with and divert audiences from The Times's original content.
Historical precedents, such as the 2013 Google Books case, established significant guidelines for "transformative use." In that case, Google’s digitisation of books for searchable content was deemed permissible. However, the current digital and AI landscapes present new challenges that may not fit neatly into past rulings. Generative AI, which creates new outputs based on large datasets, raises critical questions about whether existing fair use standards are adequate for modern technological applications.
The controversy surrounding Perplexity AI further illustrates the tension between AI companies and news publishers. Perplexity AI has faced allegations of plagiarism. In early June 2024, Forbes threatened legal action, accusing Perplexity AI of using its articles without proper attribution. Wired's investigation revealed that Perplexity AI may have used content from other major news sites without permission. The company’s ‘Pages’ feature, which generates AI-driven reports, was criticised for closely mirroring paywalled content from Forbes, raising concerns about insufficient citation. This situation highlights the challenges of ensuring proper attribution and compensation in the age of AI.
Legal Challenges: The Inadequacies of India’s Copyright Law
India’s current copyright framework, primarily governed by the Copyright Act of 1957, does not explicitly address the challenges posed by AI technologies. The concept of ‘fair dealing,’ which allows limited use of copyrighted material without permission, is insufficient in the context of AI-generated content. The Act lacks specific provisions for AI, leading to a legal grey area where the rights of content creators are not adequately protected when AI models generate commercial profits from their work. This situation has prompted calls for legal updates to address the new technological realities brought about by AI.
In a landmark decision, the Guangzhou Internet Court has held an AI company accountable for copyright infringement, marking a significant legal precedent. The court ruled that the company’s AI-generated text-to-image services, which produced images closely resembling Ultraman—an iconic character owned by Tsuburaya Productions—violated copyright laws. The court's order for the company to pay 10,000 RMB ($1,389) in damages and implement keyword filtering underscores the growing global scrutiny on AI firms regarding the unauthorized use of intellectual property. This ruling reflects a broader trend in international copyright enforcement related to AI. Meanwhile, France's competition authority fined Google €250 million ($271 million) for breaching agreements on using copyrighted content for its Bard AI service, highlighting the increasing global emphasis on enforcing copyright protections in the AI sector.
A pertinent example of the complexities involved is the ongoing lawsuit by The New York Times against OpenAI, wherein the former accuses the latter of using its copyrighted content without permission for AI training. OpenAI's defence that using publicly available content for AI training constitutes "fair use" under U.S. copyright law highlights the divergence between U.S. and Indian copyright laws. Unlike the U.S., where fair use is broader and more flexible, India’s fair dealing provisions are more restrictive, potentially leaving content creators with fewer protections. India's cautious approach to legal reform, often characterised by extensive deliberation, may struggle to keep pace with the rapid evolution of AI. The disconnect between technological advancements and the legal system could leave content creators vulnerable to exploitation, hindering the effective management of intellectual property rights in the digital age. Therefore, there is an urgent need for the Indian legal system to evolve more swiftly to address these emerging challenges.
The Ethical and Creative Dimensions of AI and Copyright
The ethical implications of generative AI are profound, particularly concerning copyright issues. Generative AI tools like OpenAI’s ChatGPT rely on vast datasets, including potentially copyrighted materials, to generate new content. This process raises significant ethical questions about the use of copyrighted content without proper attribution or compensation. The Ministry of Electronics and Information Technology (MeitY) and Indian courts are currently grappling with whether AI-generated outputs can be copyrighted and how these tools might infringe on existing copyrights.
A critical issue is the concept of “authorship,” which is central to copyright protection under Indian law. Traditionally, Indian copyright law recognises only human-created works. This principle was highlighted in 2022 when an Indian technology firm attempted to register a visual artwork created autonomously by AI. The Copyright Office rejected the application due to the lack of human authorship, signalling the need for clearer guidelines in the age of AI.
The distinction between human input and AI-generated content was further underscored in a case involving an artist seeking copyright protection for a piece created using a generative AI tool. The Copyright Office ruled that the AI-generated elements did not meet the threshold for copyrightability, indicating a trend towards differentiating between human and machine-generated works. These cases reflect a broader ethical concern about the role of AI in creative processes and the potential for AI to dilute the value of human creativity.
The global debate over AI and copyright provides valuable insights for India. For instance, the Beijing Internet Court has granted copyright protection to AI-generated content, provided there is sufficient human intellectual input. Similarly, the European Union is considering frameworks that might recognise AI-generated works under specific conditions. India, while observing these international developments, must navigate its legal landscape carefully, balancing the need for innovation with the protection of intellectual property rights.
The Role of Fair Use in AI Development: Ambiguities and Challenges
The concept of fair use, as recognized under Section 52 of the Copyright Act, could potentially play a significant role in resolving conflicts between AI development and copyright law in India. Fair use allows for the use of copyrighted material without permission for purposes such as criticism, review, or research. This provision could enable AI developers to train models on copyrighted works without infringing on the rights of original creators, provided the use meets fair use criteria.
However, the application of fair use in the context of AI remains ambiguous in India. The extent to which copyrighted material can be used before it constitutes infringement is still unresolved, leaving AI developers and content creators in a precarious position. The New York Times case against OpenAI in the U.S. offers some insights into how Indian courts might approach similar issues, though the differences in legal frameworks between the two countries suggest that India will need to develop its own approach.
The uncertainty surrounding fair use in AI development highlights a broader challenge: the need for legal clarity in the face of rapidly advancing technology. Without clear guidelines, both AI developers and content creators are likely to face continued legal disputes, potentially stifling innovation and creativity. It is imperative that India’s legal system addresses these ambiguities to foster a more balanced and fair environment for all stakeholders.
Global Perspectives on AI and Copyright: Lessons for India
India's approach to AI and copyright is increasingly shaped by global trends, reflecting a need for updated legal frameworks that can address the complex interplay between technology and intellectual property. The landscape has evolved significantly since the early days of the internet, with contemporary issues around AI and copyright drawing from both historical precedents and emerging challenges.
In 2005, the legal battle over Google's "Google Books" project became a landmark case in intellectual property law. The Association of American Publishers and the Authors Guild challenged Google's practice of scanning entire library collections without permission. The court ultimately ruled in 2013 that this scanning constituted a "transformative use" of the material, thus not infringing copyright. This case set a precedent for how digitization and transformative use are evaluated under copyright law.
Fast forward to the launch of ChatGPT and similar AI technologies, and the debate around copyright infringement has intensified. Recent high-profile legal actions, such as the lawsuit filed by The New York Times against OpenAI, underscore the ongoing tensions between AI developers and content creators. The Times alleged that OpenAI used its content without permission, further complicating the discourse around fair use and copyright infringement.
The digital ecosystem of 2024 presents a vastly different scenario from that of the early 2000s. The proliferation of user-generated content has added layers of complexity to copyright issues. Today, every piece of digital content, whether it's a TikTok dance or a smartphone photo, is automatically protected by copyright. This shift highlights the need for a nuanced approach to intellectual property that accommodates the rapid advancement of AI technologies.
Internationally, countries are taking varied approaches to address these challenges. For instance, New Zealand has introduced measures to respect Māori cultural heritage in AI applications, while Italy's Cultural Heritage Code mandates authorisation and compensation for digital uses of its artistic treasures. These examples illustrate how different jurisdictions are adapting their copyright laws to account for AI's impact on creative industries.
These international pressures are indicative of a broader global consensus on the need for updated copyright laws that reflect the realities of AI. India, as a rapidly growing hub for AI development, must navigate these pressures while considering its own unique legal and cultural context. The challenge lies in balancing the protection of content creators with the need to foster innovation in AI, ensuring that India remains competitive on the global stage.
Challenges of Content Protection and AI Advancement
In response to these challenges, publishers are exploring various strategies to combat unauthorised scraping, including rate limiting and updates to robots.txt files, which indicate which pages should not be accessed by web crawlers. For instance, Reddit is implementing measures to distinguish between AI and human traffic through updated robots.txt files and rate limiting. Despite these efforts, unauthorised scraping remains a significant challenge.
Perplexity AI has proposed revenue-sharing agreements with publishers and is considering offering its technology for interactive Q&A features on their websites. While these measures aim to address content appropriation issues, their effectiveness remains to be seen. Additionally, Cloudflare has introduced a feature to block AI developers from accessing content, utilising scoring mechanisms to detect potential bots. These initiatives reflect the ongoing struggle to balance content protection with the advancement of AI technology.
The evolving landscape of AI and copyright law underscores the need for clearer legal and technical standards to protect intellectual property while fostering innovation. As AI technologies advance, establishing robust frameworks to address fair use, plagiarism, and web scraping protocols is crucial. The ongoing debates and legal battles will shape how content creators and AI companies navigate the future of digital content and copyright law, striving to balance creative protection with the advancement of AI technology.
New Approaches to Content Licensing and Revenue
As the digital landscape evolves, new monetisation models are emerging to address the challenges posed by AI and unauthorised content use. Tollbit, a startup that secured $7 million in March 2024, exemplifies these innovations. Tollbit seeks to streamline content monetisation by facilitating direct licensing agreements between publishers and AI companies. The platform operates like a "tollbooth" on the internet, enabling web scrapers and AI agents to pay for content use based on terms set by the publishers. This cooperative model offers flexible agreements, including bulk rates for extensive content use, potentially simplifying the tracking of content usage and ensuring fair compensation.
Despite the promise of Tollbit's model, its success hinges on several key assumptions. AI companies must be willing to invest in quality content and prioritise diverse sources over cheaper alternatives. There is a risk that AI systems may favour less expensive content, potentially undervaluing premium publishers.
As India navigates its growing AI sector, the legal and ethical implications of AI-generated content and the use of copyrighted material necessitate careful consideration. The existing copyright framework is inadequate for addressing these contemporary issues, highlighting the need for legal reforms that balance the protection of content creators with the promotion of innovation. The evolving debates around fair compensation, the nuances of fair use, and the rise of new monetisation models all underscore the complex interplay between AI and copyright. By drawing on global perspectives and tackling these challenges head-on, India has the opportunity to lead in developing a more equitable and innovative digital landscape.
IP Round-up
Rs 20 Lakhs Fine for Music Copyright Breach:
The Delhi High Court has fined Rakshit Shetty, the renowned Kannada actor and owner of Paramvah Studios, Rs 20 lakhs for unauthorised use of music tracks from MRT Music in his film Bachelor Party. MRT Music, accused Shetty of using songs from Nyaya Ellide and Gaalimaathu without permission. Despite an open letter posted by Shetty on Instagram claiming MRT Music's fees were excessive, he and his studio failed to attend court, leading to the fine and a directive to remove the social media post. (Source: Deccan Chronicle)
Burger King Corp Loses Trademark Battle in Pune:
In a landmark ruling, Pune District Judge Sunil Vedpathak dismissed a 13-year-old lawsuit filed by Burger King Corporation against a local eatery named 'Burger King.' The US fast-food giant had sought a permanent injunction and Rs 20 lakh in damages, alleging trademark infringement. However, the court found that the Pune-based restaurant had been operating under the name since 1991-92, well before Burger King Corporation entered India in 2014. The judge ruled that the plaintiff failed to demonstrate customer confusion or infringement by the local eatery. (Source: Business Standard)
Jan Vishwas Act: A New Era for Trademark Enforcement in India: The Jan Vishwas (Amendment of Provisions) Act, 2023, effective from August 1, 2024, brings significant changes to the Indian Trade Marks Act, 1999. Key amendments include replacing imprisonment with monetary penalties under Section 107(2) and eliminating Sections 106, 108, and 109. The Act introduces new adjudication and appeal mechanisms through Sections 112A and 112B, requiring the appointment of adjudicating officers and appellate authorities. These reforms aim to streamline enforcement and compliance, but may also present new administrative challenges for the Trademarks Registry and stakeholders adjusting to the updated system. (Source: Lexology)
AstraZeneca Wins Overturned Verdict in Patent Dispute: AstraZeneca has successfully overturned a jury's $107.5 million award for patent infringement involving Pfizer's cancer-drug patents. On August 16, 2024, U.S. District Judge Matthew Kennelly ruled that Pfizer’s patents related to Tagrisso, AstraZeneca’s lung-cancer drug, were invalid due to insufficient written descriptions. Pfizer, which acquired Wyeth in 2009, had claimed Tagrisso infringed patents linked to the breast-cancer drug Nerlynx. Despite the initial jury decision in May, Kennelly found the patents did not enable reproduction of the inventions. AstraZeneca’s Tagrisso generated nearly $5.8 billion in revenue last year. (Source: Medical Dialouges)
Fitabeo Therapeutics Secures Japanese Patent for Innovative Drug Delivery Tech: Fitabeo Therapeutics, a clinical-stage pharmaceutical company spun out from the University of Strathclyde, has been granted a Japanese patent for its advanced drug delivery technology, SpherionexT. This controlled-release film technology aims to enhance the efficacy, safety, and patient adherence of therapeutic agents by improving drug release precision. The University’s Director of Innovation, Meryl Levington, expressed optimism about expanding patent coverage globally to attract investors and expedite new medicine development. Chief Scientific Officer, Professor Alex Mullen, emphasized the technology's potential to revolutionize treatment standards and improve patient outcomes, reflecting the team's commitment to pharmaceutical innovation. (Source: University of Strathclyde)
References:
Bellan, R. (2024, July 2). News outlets are accusing Perplexity of plagiarism and unethical web scraping. TechCrunch. https://techcrunch.com/2024/07/02/news-outlets-are-accusing-perplexity-of-plagiarism-and-unethical-web-scraping/
Business Today. (2024, April 5). Indian govt to frame new AI law to protect rights of news publishers, content creators: Report. Business Today. https://www.businesstoday.in/technology/news/story/indian-govt-to-frame-new-ai-law-to-protect-rights-of-news-publishers-content-creators-report-424394-2024-04-05
Costigan, J. (2024, March 1). China rules AI firm committed copyright infringement. Forbes. https://www.forbes.com/sites/johannacostigan/2024/02/29/china-rules-ai-firm-committed-copyright-infringement/
Daws, R. (2024, April 29). FT and OpenAI ink partnership amid web scraping criticism. Artificial Intelligence News. https://www.artificialintelligence-news.com/news/ft-and-openai-ink-partnership-web-scraping-criticism/
Gold, J. (2024, March 20). French regulator fines Google $271M over generative AI copyright issue. CIO. https://www.cio.com/article/2069449/french-regulator-fines-google-271m-over-generative-ai-copyright-issue.html
Hasson, E. (2024, February 20). The New York Times vs. OpenAI: A turning point for web scraping. Imperva. https://www.imperva.com/blog/the-new-york-times-vs-openai-a-turning-point-for-web-scraping/
Kenneally, C. (2024, May 23). That’ll cost you, ChatGPT: Copyright needs an update for the age of AI. The Hill. https://thehill.com/opinion/technology/4670494-thatll-cost-you-chatgpt-copyright-needs-an-update-for-the-age-of-ai/
Mahadik, K. (2024, July 4). Perplexity AI controversy: News publishers' allegations of copyright. The Indian Express. https://indianexpress.com/article/technology/tech-news-technology/perplexity-ai-controversy-news-publishers-allegations-copyright-9428935/
Motegi, Y. (2024, July 18). Japan newspaper association calls for legal reform to curb AI search. Nikkei Asia. https://asia.nikkei.com/Business/Technology/Artificial-intelligence/Japan-newspaper-association-calls-for-legal-reform-to-curb-AI-search
Music Plus. (2024, February 14). Copyright clarity in India: AI developers require permission for commercial AI models. Music Plus. https://www.musicplus.in/copyright-clarity-in-india-ai-developers-require-permission-for-commercial-ai-models/
NeoMam Studios. (2023, December 12). Mapped: Interest in generative AI by country. Visual Capitalist. https://www.visualcapitalist.com/cp/mapped-interest-in-generative-ai-by-country/
Obhan & Associates. (2024, August 16). China's AI Surge: Lessons for India's Tech Future. Lexology. https://www.lexology.com/library/detail.aspx?g=8e184ea4-f73e-403e-930d-863298f83d7d
Peck, G. A. (n.d.). News publishers vs. generative AI: Can copyright law keep up? Editor & Publisher. https://www.editorandpublisher.com/stories/news-publishers-vs-generative-ai-can-copyright-law-keep-up,251487
PTI. (2024, August 18). US-based Burger King loses infringement suit against namesake Pune eatery. Business Standard. https://www.business-standard.com/companies/news/us-based-burger-king-loses-infringement-suit-against-namesake-pune-eatery-124081800175_1.html
Roy, A. (2024, January 26). Indian publishers seek rules for copyright protection against generative AI models. The Economic Times. https://economictimes.indiatimes.com/tech/technology/indian-publishers-seek-rules-for-copyright-protection-against-generative-ai-models/articleshow/107154425.cms?from=mdr
Roy, T. L. (2024, January 5). Indian publishers seek legal recourse against content misuse in free-for-all AI world. Storyboard18. https://www.storyboard18.com/quantum-brief/indian-publishers-seek-legal-recourse-against-content-misuse-in-free-for-all-ai-world-19680.htm
Scannell, E. (2024, June 14). AI, copyright and fair use: What you need to know. TechTarget. https://www.techtarget.com/searchenterpriseai/feature/AI-copyright-and-fair-use-What-you-need-to-know
Schiffrin, A., & Mateen, H. (2024, May 14). Startup aims to help publishers collect fees from AI companies. Tech Policy Press. https://www.techpolicy.press/startup-aims-to-help-publishers-collect-fees-from-ai-companies/
Sharma, B. (2024, August 18). Bachelor Party controversy: Rakshit Shetty to shell out huge amount. Deccan Chronicle. https://www.deccanchronicle.com/entertainment/bachelor-party-controversy-rakshit-shetty-on-shell-out-huge-amount-1817390
Sharma, R. (2024, August 19). AstraZeneca gets USD 107 million Pfizer verdict overturned in US cancer drug patent fight. Medical Dialogues. https://medicaldialogues.in/news/industry/pharma/astrazeneca-gets-usd-107-million-pfizer-verdict-overturned-in-us-cancer-drug-patent-fight-133379
Singh, C. (2024, January 6). What Indian media can do to tackle misuse of content by AI. NDTV Profit. https://www.ndtvprofit.com/law-and-policy/what-indian-media-can-do-to-tackle-misuse-of-content-by-ai
Statista. (n.d.). Generative AI market size worldwide. Statista. https://www.statista.com/outlook/tmo/artificial-intelligence/generative-ai/worldwide#market-size
Takeuchi, S., & Mashiko, H. (2024, August 20). News media groups warn of free-riding by AI sites, lose views as AI imitates copyrighted content. Asia News Network. https://asianews.network/news-media-groups-warn-of-free-riding-by-ai-sites-lose-views-as-ai-imitates-copyrighted-content/
The Letter Two. (2024, July 5). AI economy: Publishers push back over AI platforms' content use. The Letter Two. https://thelettertwo.com/2024/07/05/ai-economy-publishers-push-back-over-ai-platforms-content-use/
University of Strathclyde. (2024, August 20). University spin-out secures patent for innovative drug delivery tech. https://www.strath.ac.uk/whystrathclyde/news/2024/universityspin-outsecurespatentforinnovativedrugdeliverytech/
Venditti, B. (2024, August 6). Ranked: Top companies by generative AI patents. Visual Capitalist. https://www.visualcapitalist.com/ranked-top-companies-by-generative-ai-patents/
