Introduction
India has taken a definitive step toward establishing a robust legal foundation for Artificial Intelligence (AI) and copyright governance. On December 8, 2025, the Department for Promotion of Industry and Internal Trade (DPIIT) released Part I of a landmark Working Paper. Prepared by a specialised expert committee, this document outlines the government’s inaugural strategy for regulating the use of copyright-protected content in AI training, inviting public feedback within a 30-day window. The proposal centers on a mandatory blanket licensing regime. This would grant AI developers the right to utilise all “lawfully accessed” copyrighted material for training purposes, balanced by a statutory right to remuneration for creators. It is a hybrid model designed to resolve the friction between accelerated AI development and the preservation of human creative rights.
The Necessity of Regulatory Intervention
Existing copyright laws in India do not specifically address the nuances of machine learning. Generative AI models require vast datasets comprising literature, imagery, and audio often sourced through web scraping or third-party vendors. This has triggered a global debate over:
- Unlicensed Reproduction: The use of protected works without explicit consent.
- Economic Impact: Potential loss of livelihood for human authors and artists.
- Output Risks: The possibility of AI generating content that mirrors protected works.
- Market Inefficiency: The prohibitive cost and complexity of negotiating individual licenses for millions of data points.
In response, the DPIIT formed an eight-member committee on April 28, 2025, led by Additional Secretary Himani Pande. Their mission was to evaluate the adequacy of current laws and propose reforms that modernize India’s intellectual property landscape.
The Core Recommendation
The Committee’s majority view favors a streamlined licensing model with four pillars:
- Lawful Access Threshold: Access is granted for any work obtained legally, including purchased media, authorized databases, and publicly available online content.
- Compulsory Participation: To prevent data monopolies and bottlenecks, copyright holders cannot “opt out” of the training regime.
- Statutory Compensation: In exchange for the loss of exclusive control, creators receive mandatory royalties managed by a government-approved “umbrella organization.”
- Centralized Interface: AI companies particularly startups and MSMEs benefit from a “single-window” system to clear rights efficiently.
The committee recognizes a dual imperative to harmonise the Technological Necessity of frictionless data access without which AI development becomes unviable due to the burden of negotiating millions of individual contracts with Creative Equity, ensuring creators are compensated for the intellectual efforts powering commercial AI. By addressing the high “transaction costs” that otherwise stifle domestic innovation and disadvantage Indian startups against global giants, the proposed framework offers five strategic advantages: Data Certainty through guaranteed access to training materials, Efficiency in reducing administrative hurdles, Judicial Safeguards to ensure fair royalty structures, Simplified Compliance via centralized collection, and Democratic Access to create a level playing field for smaller innovators.
Impact on Stakeholders
- For Creators: The benefit is guaranteed revenue. While they lose the power to exclude their works, they gain a structured, transparent stream of royalties that was previously non-existent in the “scraping” era.
- For AI Developers: The model offers legal immunity and operational speed. It replaces the threat of litigation with a clear, predictable pathway for data usage.
- For Startups and MSMEs: By lowering the “barrier to entry,” the framework fosters a more competitive and inclusive technological ecosystem.
- For Regulators: The proposal necessitates enhanced oversight. Authorities will need to establish precise guidelines for royalty distribution and modernize existing copyright rules to match this new reality.
Navigating Legal and Policy Hurdles
Despite its progressive nature, several challenges remain:
- International Treaties: Ensuring the plan complies with the “Three-Step Test” under the Berne Convention and TRIPS.
- Economic Valuation: Determining the exact value of data in a way that is fair to creators but not bankrupting for startups.
- Definition of “Lawful”: Clarifying the status of cached data and user-generated content.
- Output Integrity: Ensuring AI models do not generate infringing “copycat” content.
Conclusion
This Working Paper represents a foundational shift in India’s digital policy. It acknowledges that for India to become a global AI superpower, it must create a frictionless environment for data while providing a safety net for its creative economy. As the consultation period begins, the feedback from publishers, tech firms, and legal experts will refine this blueprint. Ultimately, this framework will define how India balances the tools of the future with the protections of the past.
Expositor(s): Adv. Archana Shukla
