Understanding India's 'One Nation, One Licence' Framework

Understanding India's 'One Nation, One Licence' Framework

The rapid advancement of Artificial Intelligence (AI) and Large Language Models (LLMs) such as ChatGPT has raised significant concerns regarding the use of copyrighted content for AI training without appropriate consent or remuneration. This issue has led to a surge in global litigation, extensive policy discussions, and an atmosphere of regulatory uncertainty, particularly at the intersection of technology, intellectual property rights (IPR), innovation, and regulation. The role of the State in rate regulation and compulsory licensing has become increasingly crucial.

In this context, a committee led by the Department for Promotion of Industry and Internal Trade (DPIIT) has released a working paper proposing a statutory licensing framework. This framework aims to create a balance between AI innovation and copyright protection in India.

Key Proposals

The main proposal is the concept of 'One Nation, One Licence, One Payment'. This mandates that all AI developers must pay royalties for the use of copyrighted works in their AI training processes. Notably, there is no opt-out mechanism for content that is freely available online. This model is inspired by the compulsory licensing provisions in Indian copyright law, particularly as they apply to radio broadcasting.

The committee has explicitly rejected the notion of voluntary licensing agreements, such as the ones between OpenAI and the Associated Press. The reasons for this rejection include high transaction costs, unequal bargaining power, and the marginalization of smaller creators and startups. The committee believes that voluntary licensing disproportionately favors large technology companies and major publishers.

Institutional Mechanism – CRCAT

A new umbrella non-profit organization, the Copyright Royalties Collective for AI Training (CRCAT), is proposed to be established under the Copyright Act of 1957. This body will be responsible for collecting royalties from AI companies and distributing the proceeds among copyright holders. Membership will be restricted to organizations, not individuals, with one member per class of work, and coverage may gradually expand to include unorganized sectors.

Royalty Determination Framework

A government-appointed rate-setting committee will be responsible for determining fair and transparent royalty rates. This committee will include senior government officials, legal experts, economic and financial specialists, AI and emerging technology professionals, and representatives from AI developers and CRCAT. They will have the authority to fix predictable rates and review them every three years, with their decisions subject to judicial review. Initially, a flat rate may be preferred, based on a percentage of the gross global revenue generated from commercialized AI systems, excluding taxes.

Retroactive Application of Royalties

Royalties will apply retroactively, meaning that AI developers who have already utilized copyrighted works and generated revenue will be required to pay past dues. This approach aims to ensure fairness and accountability, serving as a corrective measure to restore balance in the creative ecosystem.

Transparency and Burden of Proof

AI developers will be mandated to disclose data, submitting a 'Sufficiently Detailed Summary' of the datasets utilized. This summary must include details about the type of data (text, images, music, audiovisual), the source (social media, publications, libraries, public datasets, proprietary data), and the nature of data usage. The CRCAT will distribute royalties based on the extent of usage, with heavily used categories such as news, music, and audiovisual content receiving larger shares.

In legal disputes, a legal presumption will exist; content owners will need to prove that their claims are valid, shifting the burden onto AI developers to demonstrate misuse or non-payment.

Stakeholder Responses

Supporters of the proposal argue that it ensures non-discriminatory access to training data, prevents the concentration of royalties among a few large players, and creates a predictable legal environment for AI development. However, opponents, including NASSCOM, label the mandated royalties as a "tax on innovation" and advocate for opt-out mechanisms for content creators. Concerns within the creative industry highlight the unprecedented nature of government-fixed rates, fearing that this could lead to the undervaluation of premium content.

Challenges and the Way Forward

There are risks associated with over-regulation potentially stifling AI innovation, emphasizing the need for robust stakeholder consultations. The administrative complexity involved in royalty distribution necessitates careful adjustments to royalty rates to avoid discouraging AI startups. Resistance is anticipated from both AI firms, due to cost burdens, and content creators, who may oppose flat-rate structures. Strong judicial oversight will be essential to prevent arbitrary decision-making and ensure India does not become an outlier in global AI copyright regulation.

In conclusion, India's proposed mandatory blanket licensing regime for AI training represents a bold and interventionist strategy to reconcile innovation with copyright protection. By institutionalizing royalty payments through a statutory mechanism, the Centre aims to ensure equitable compensation for creators while allowing open access to training data for AI developers. The success of this model will hinge on the rationality of rates, transparency in operations, and adaptive governance, making it a critical test case for AI regulation in the Global South.

Frequently Asked Questions (FAQs)

Q1. What is the 'One Nation, One Licence' framework?
Answer: It is a proposed statutory licensing framework in India that mandates all AI developers to pay royalties for using copyrighted content in AI training, aimed at balancing innovation and copyright protection.

Q2. Who will manage the royalty collection?
Answer: A new non-profit organization called the Copyright Royalties Collective for AI Training (CRCAT) will be established to handle royalty collection and distribution among copyright holders.

Q3. What are the implications for AI developers?
Answer: AI developers will need to pay royalties retroactively for previously used copyrighted works and submit detailed data usage summaries to ensure compliance with the new framework.

Q4. How will royalty rates be determined?
Answer: A government-appointed committee will set fair and transparent rates, with reviews every three years, ensuring that rates reflect the true value of the copyrighted content.

Q5. What challenges does the framework face?
Answer: Potential over-regulation, administrative