Quick Overview
India is considering a major overhaul of the Copyright Act, 1957 through a hybrid model framework to address copyright challenges posed by Generative Artificial Intelligence (AI). The proposal introduces a mandatory blanket licence for AI training along with statutory remuneration for creators, aiming to balance technological innovation with intellectual property protection.
Introduction
The rapid expansion of Generative Artificial Intelligence has fundamentally altered how creative content is produced, consumed, and monetised. AI systems rely heavily on vast quantities of copyrighted material for training, raising complex legal questions regarding authorship, consent, compensation, and enforcement. India’s Copyright Act, 1957—drafted long before the digital age—lacks clarity on these emerging challenges. In response, the government has proposed a hybrid model framework that seeks to harmonise innovation with creators’ rights while positioning India distinctly from global regulatory approaches.
Why in News?
The Indian government has initiated consultations to amend the Copyright Act, 1957 to address AI-related copyright concerns. The proposed hybrid model introduces mandatory licensing for AI training and statutory remuneration for rights holders, marking a significant shift from the traditional permission-based copyright regime.
What is the Hybrid Model Framework?
The hybrid model combines two key elements:
Mandatory Blanket Licensing for AI developers to use copyrighted works for training.
Statutory Right to Remuneration for creators whose works are used, regardless of consent.
This approach prioritises innovation while ensuring creators receive fair compensation.
Proposed Amendments to the Copyright Act, 1957
1. Blanket Licence for AI Training
AI developers would receive a non-negotiable, universal licence to use lawfully accessed copyrighted content for training.
Rights holders would lose the right to withhold consent for AI training.
This departs from Section 14 of the Act, which currently grants exclusive control to copyright owners.
2. Statutory Remuneration Rights
Copyright holders would be entitled to royalties as a fixed percentage of AI-generated revenue.
Royalties would be payable only after commercialisation of AI models.
Consent is not required; compensation is guaranteed by statute.
3. Centralised Collection Mechanism
A government-designated non-profit entity would collect royalties.
Copyright societies and Collective Management Organisations (CMOs) would manage distribution.
This ensures transparency and reduces litigation.
4. Protection for Startups and MSMEs
Differential royalty structures or concessions may be introduced.
This prevents excessive compliance costs from stifling innovation.
Why is Amendment of the Copyright Act Necessary?
Outdated Legal Framework
The Act lacks definitions for:
Artificial Intelligence
Generative AI
Machine Learning
Training data
This creates legal uncertainty in applying existing provisions.
Ambiguity on AI Training
It is unclear whether AI training amounts to reproduction or infringement under Section 14.
Applicability of fair dealing exceptions under Section 52 remains unsettled.
Unauthorized Use of Copyrighted Works
AI models are trained through mass scraping of copyrighted material without permission or payment.
This undermines the consent-based structure of copyright law.
Authorship of AI-Generated Works
Section 2(d)(vi) defines authorship only in human terms.
Section 2(ffc) treats computers as tools, creating confusion over ownership of AI-generated outputs.
Cross-Border Enforcement Challenges
Section 40 focuses on territorial reciprocity but fails to address:
Cloud-based AI training
Cross-border data flows
AI systems trained abroad using Indian content
Judicial and Global Perspectives
Indian Courts
Courts have upheld automatic copyright protection and fair dealing principles.
Recent judgments emphasise proactive enforcement against digital piracy.
Global Approaches
USA: Requires human creativity for copyright protection.
EU: Focuses on training data transparency under the AI Act, 2024.
China: Recognises AI-generated works where human input ensures originality.
UK: Assigns copyright to the person making “necessary arrangements” for computer-generated works.
India’s hybrid model offers a distinct middle path between strict consent regimes and unrestricted AI training.
Conclusion
The proposed hybrid model framework represents a pragmatic response to the challenges posed by Generative AI. By combining mandatory licensing with statutory remuneration, India seeks to safeguard creators’ economic interests without obstructing innovation. However, effective implementation, transparent royalty distribution, and international coordination will be critical to ensuring that the reforms strengthen India’s digital economy while preserving the integrity of intellectual property rights.
CLAT / UPSC Exam Relevance Summary
Prelims: Copyright Act, 1957, AI, Generative AI, fair dealing, Section 14 & 52
GS Paper 3: Intellectual Property Rights, emerging technologies, innovation policy
GS Paper 2: Governance, regulatory reforms
CLAT: Technology law, copyright, public policy implications