AI–copyright Overhaul with ‘One Nation, One Licence, One Payment’ Model

India's AI-Copyright Licensing Reform: Bold Intent, Fragile Execution

On December 10, 2025, the Department for Promotion of Industry and Internal Trade (DPIIT) unveiled its working paper proposing a landmark reform titled ‘One Nation, One License, One Payment: Balancing AI Innovation and Copyright.’ The draft envisions India becoming the first country to mandate a statutory licensing regime for artificial intelligence (AI) developers, linking royalties for creators directly to AI training datasets. At stake lie the rights of copyright holders, India’s ambitions for sovereign AI capabilities, and the inevitable tension between government fiat and market freedom.

Mandatory licensing, the creation of a centralized Copyright Royalties Collective for AI Training (CRCAT), and retroactive royalty payments are key pillars of the proposed framework. The model rejects voluntary licensing deals like the OpenAI-AP settlement seen overseas, citing unequal bargaining power and monopolistic risks. Instead, royalty rates will be set by a government-appointed committee under the Copyright Act, 1957, and adjusted every three years to ensure fairness and reflect technological changes. On paper, the framework promises to inject equity into an uneven playing field dominated by AI multinationals. The question is whether DPIIT’s vision can survive the practical pitfalls of enforcement, industry pushback, and judicial oversight.

Architectural Details: A Blueprint for Friction

Three structural props support the proposed licensing overhaul:

• Statutory Blanket Licensing: AI developers will be required to obtain a license granting access to copyrighted material for training purposes. Any bypass of paywalls or unauthorized downloads will violate the provisions.
• CRCAT: An umbrella institution established under the Copyright Act, 1957, tasked with collecting royalties and distributing them among classes of copyright holders.
• Royalty Setting Committee: This government-appointed body—comprising legal, economic, and technological experts alongside CRCAT members—will set tariff rates that are transparent, scalable, and adaptable every three years.

No less ambitious is the push for retroactive payments, which aims to hold AI companies accountable for historical extraction of copyrighted datasets. While legally justified, retroactivity could trigger cascading lawsuits, with copyright holders bearing the burden of proving misuse. As an institution-heavy proposal resting on judicial oversight and regulatory discipline, the framework risks becoming a bureaucratic swamp rather than an innovation enabler.

Fragility Beneath Bold Innovation

The headline commitment—to level the playing field for copyright holders while safeguarding AI innovation—is laudable but fraught with risk. The decision to reject voluntary licensing agreements echoes a valid distrust of monopolistic concentration. High-profile deals like OpenAI’s arrangement with news outlet AP have shown that tech giants often wield disproportionate power over content creators. However, mandatory licensing risks imposing a one-size-fits-all rate structure, potentially stifling innovation in nascent startups unable to absorb licensing costs.

The hybrid compensation model is another site of contention. While creators are guaranteed royalties by law, industries housing diverse creative economies—the publishing sector, for instance—are likely to see disputes over revenue allocation within the CRCAT. DPIIT suggests that royalty rates will remain “fair” and scaled, but leaves opaque how smaller creators or marginalized creative ecosystems will fare under blanket royalty distribution.

Moreover, operational funding for the CRCAT remains undefined. Will the umbrella body risk becoming the AI equivalent of India’s GST Council—effective but perennially trapped in inter-group friction and funding quarrels? The working paper addresses technical feasibility but largely skirts institutional clarity.

Structural Fault Lines: Scale, Entry Barriers, and Judicial Bottlenecks

A deeper concern lies in multi-dimensional friction—from industry-scale dependency to judicial intervention risks:

• Startups vs Giants: While India’s AI sector has grown by 48% globally (2025 estimates), statutory royalties risk favoring multinational incumbents over emerging firms. Small companies may avoid using copyright-protected data altogether to cut costs, limiting their competitive edge.
• Judicial Bottleneck: By allowing royalty rates to be challenged in court, DPIIT essentially invites litigation-heavy delays. Cases could linger across multiple jurisdictions, as parties dispute retroactive liability for AI systems stretching back years.
• Overcentralized Governance Risk: The CRCAT structure places sweeping economic control in one institutional basket. In democracies like India, high-centralization frequently spawns inefficiency and rent-seeking rather than equitable arbitration.

This is part of a broader trend of Indian policy models—ambitious architectures saddled by weak implementation. Much depends on inter-ministerial coordination between DPIIT, the Ministry of Electronics and Information Technology (MeitY), and enforcement agencies. A fragmented approach could undermine the very ideals of equitable access and accountability.

Lessons from Australia: A Contrast in Approach

Globally, regulatory experiments in AI copyright have leaned toward decentralized, sector-driven systems rather than top-down mandates. Australia’s News Media Bargaining Code, for instance, requires platforms like Google and Facebook to negotiate payments directly with news publishers, factoring sectoral diversity and bargaining asymmetries into granular agreements. While this voluntary, localized structure avoids the pitfalls of overcentralization, its scalability to AI ecosystems remains untested.

By contrast, India’s statutory licensing model pursues legal rigidity at the risk of excluding context-sensitive agreements altogether. The divergence underscores philosophical tensions between control and flexibility—not abstract trade-offs but real, structural consequences.

The Road Forward: Metrics of Success and Lingering Challenges

For DPIIT’s licensing model to succeed, enforcement cannot merely rely on centralized compliance mechanisms. Success hinges on:

• Clear implementation timelines for the CRCAT to avoid bureaucratic delays.
• Transparent royalty-setting processes involving multi-stakeholder consultation.
• Operational frameworks balancing industry needs with creator equity—scalable and sector-responsive.

Metrics to watch include domestic AI adoption rates post-reform, retention of startups within the ecosystem, and judicial precedents emerging from inevitable retroactive payment disputes. Yet large unresolved challenges loom, from industry resistance to entry barriers that disproportionately hurt smaller players.

Practice Questions for UPSC

Prelims Practice Questions

Frequently Asked Questions