A Relic in the Age of AI, Struggling To Protect Creators

Is India’s copyright law still relevant in an era where artificial intelligence (AI) can create, curate, and repurpose content at an unprecedented scale? Can legislation designed in 1957 protect journalists, writers, and content creators from the voracious appetite of AI-driven corporations? Or is it time to rewrite the rules before creative industries are cannibalized by technology giants?

The lawsuit filed by ANI against OpenAI in the Delhi High Court is not just another intellectual property dispute — it is a battle that could define the future of content ownership in India. At stake is a fundamental question: can generative AI companies freely consume copyrighted material to train their models without compensating the original creators?

This is not an isolated concern. Across the world, media publishers and content creators are taking legal action against AI firms for scraping their copyrighted material. From The New York Times’ lawsuit against OpenAI in the United States to Europe’s tightening copyright laws, global players are reacting to the AI revolution with urgency. But India? India’s legal framework remains shackled to an outdated Copyright Act that is ill-equipped to deal with the realities of AI.

A Law from the Past, Facing the Future

India’s Copyright Act of 1957, drafted long before the digital age, was designed to protect authors, artists, and publishers from unauthorized reproduction and distribution of their work. However, it was never intended to deal with AI models that indiscriminately harvest data from the internet to generate human-like text, images, and even code.

In its current form, Indian copyright law grants creators exclusive rights over their content, making unauthorized reproduction a clear infringement. The law recognizes that original works — whether news articles, literature, or music — require significant investment and effort, and thus must be protected from free-riders. However, AI models operate in a legal gray zone, scraping vast amounts of publicly available information and using it to generate content that is strikingly similar to copyrighted works.

OpenAI claims that ChatGPT only uses “publicly available data,” but this defense is deeply flawed. Just because content is accessible online does not mean it is free to use. News publishers and journalists, for example, spend enormous resources on reporting, editing, and fact-checking — only for AI models to repurpose their work without attribution or compensation.

This creates a legal paradox:

• AI companies are violating copyright law by using protected content to train their models.
   
• However, the AI-generated output is often “original” and does not directly copy the material, making it harder to classify as an infringement.

How the World is Moving Ahead — But India is Falling Behind

While India’s laws remain frozen in time, other nations have already begun to adapt.

• United States: The U.S. follows a broad “fair use” doctrine that allows unlicensed use of copyrighted material under certain circumstances. The landmark Google Books case (2015) ruled that large-scale text and data mining for research purposes does not necessarily infringe copyright.
   
• European Union: The EU has stricter laws, requiring AI companies to seek explicit permission before using copyrighted works for training their models. 
   
• Singapore: In 2021, Singapore proactively amended its copyright laws to allow text and data mining for computational analysis, striking a balance between innovation and intellectual property rights.

India, however, has no such provisions. Section 52 of the Indian Copyright Act lists exceptions for fair use, such as research, criticism, education, and news reporting, but it does not account for large-scale AI training. Without legislative intervention, Indian courts are left grappling with a law that is fundamentally unequipped to handle AI-driven copyright violations.

The Dangers of a Weak Copyright Regime

The inability to regulate AI’s usage of copyrighted content is not just a legal issue — it is a threat to India’s creative economy.

1. Journalism & Media at Risk
   If AI companies can freely use news articles to train their models, why would users visit news websites when they can get AI-generated summaries instead? Media houses could lose revenue, making independent journalism unsustainable.
    
2. Unfair Competition
   Imagine an Indian content creator spending months researching and writing an article, only for ChatGPT to summarize and rephrase it instantly without credit. AI models are essentially "reaping where they have not sown," unfairly competing with human creators.
    
3. Legal Limbo for Copyright Holders
   If a newspaper article is used to train an AI model, who owns the rights to the AI-generated content? Is it the publisher, the AI company, or the user generating the content? The lack of legal clarity creates an existential crisis for authors, journalists, and artists.
    
4. India’s AI Leadership at Stake
   India aspires to be a global AI powerhouse, but weak copyright protections could stifle innovation rather than accelerate it. If content creators are not incentivized to produce high-quality work, AI models will suffer from poor training data, diminishing India’s competitiveness in the AI race.

The Way Forward: A Legal Overhaul for The AI Age

The ANI vs. OpenAI case underscores an urgent need for India to modernize its copyright laws. Here’s how it can be done:

1. Introduce AI-Specific Copyright Provisions

• Amend Section 52 to explicitly address AI training data.
   
• Require AI firms to obtain licenses from content creators before using their material.

Implement a Compensation Framewor

• AI companies profiting from copyrighted material should pay royalties to content owners.
   
• Develop a collective licensing model, similar to how music royalties are handled, ensuring fair compensation for journalists, writers, and publishers. 

Strengthen Fair Use Guidelines

• Define clear boundaries for AI’s use of copyrighted content.
   
• Distinguish between AI models used for research (which may be permitted) and commercial AI products (which should require licenses). 

Expand Unfair Competition Laws

• Allow courts to consider "unfair competition" claims against AI companies that exploit copyrighted content.
   
• Prevent tech firms from monopolizing AI-generated knowledge at the expense of original creators.

Encourage Transparency in AI Training

• Require AI firms to disclose their training datasets and ensure they have obtained rights to the material used.
   
• Mandate attribution so that AI-generated content credits the original sources.

The Future of Copyright in the AI Era

The battle between OpenAI and ANI is not just a legal dispute — it is a turning point in India’s intellectual property landscape. If India fails to act, AI companies will continue to exploit creators, siphon revenue from publishers, and erode the foundation of the creative economy.

Will India remain stuck in outdated copyright laws while the rest of the world moves forward? Or will it recognize the urgency of the moment and enact bold reforms to safeguard its creative industries?

The choice is clear. If India wants to be a leader in AI, it must also lead in protecting those who create the content that powers AI.

Image (c) istock.com