NFTS, Generative AI And Copyright In India - Re- Examining Authorship, Originality And Infringement Tests
Author: LawSphereX
Category: Technology, Cyber & Privacy
Published:
Updated:
NFTS, GENERATIVE AI AND COPYRIGHT IN INDIA: RE-
EXAMINING AUTHORSHIP, ORIGINALITY AND
INFRINGEMENT TESTS
Sanvi Singh, Symbiosis Law School Pune
ABSTRACT
The Indian copyright system faces increasing demands to manage creative works and commercial activities which have emerged after its original design because the system must now handle blockchain-based non-fungible tokens and AI-based generative systems and human-machine collaborative artworks. The Copyright Act of 1957 protects original literary and dramatic and musical and artistic works through Section 131 while Section 2(d)2 establishes computer-generated works as protected creations which identify the creator as the person who makes the work actual.
The current digital practices raise three main questions which existing legal texts fail to answer because they lack specific definitions about who becomes an AI-enabled author and which originality standard should apply to works created through multiple production processes and what infringement doctrine should determine when copyright protected material gets used for minting and model creation and output generation and public distribution without permission. The Indian legal system needs to change existing legal principles through a complete shift in practice instead of this complete transformation of current legal standards. The primary purpose of NFTs exists to function as transaction-based digital assets which only transfer copyright ownership when users fulfil statutory written assignment requirements.
The evaluation of generative AI outputs requires a human-cantered investigation of authorship which divides between two types of creative control because originality must follow the Indian standard of skill and judgment rather than machine-generated originality3. Courts need to divide infringement cases into three categories which involve training inputs and model outputs and tokenized dissemination because each category involves distinct exclusive rights under Section 14 and requires different evaluation of substantial reproduction and public communication and intermediary accountability.
1 The Copyright Act 1957, section 13
2 The Copyright Act 1957, section 2d
3 Ownership of AI Generated Content: A Deep Dive into Copyright Law in India’ (Khurana & Khurana, 24
October 2025)
Keywords: NFT, generative AI, copyright, India, authorship, originality, infringement, computer-generated works.
I. Introduction
Copyright protection extends to original literary works, dramatic works, musical works, artistic works, cinematograph films, and sound recordings according to Section 134. Section 145 defines copyright as an exclusive bundle of rights that includes reproduction in any material form, electronic storage, issuing copies to the public, communication to the public, adaptation, and, in relation to artistic works and films, electronic storage and public communication. The provisions of this law extend their reach to all activities that involve digital copying and online display and NFT-linked distribution and machine-readable work storage. Section 2(d)6 is crucial because it defines “author” differently across categories of works. The author of a literary or dramatic work represents the author, while the composer represents musical works.
The author of artistic works other than photographs is an artist. The photographer represents the author of photographic works. The producer represents the author of films and sound recordings. The author of any literary dramatic musical or artistic work that is computer-generated is defined as the person who creates the work. This final phrase provides a statutory foothold for analysing AI-assisted creativity, but it does not itself specify how much human control is required. The rules that determine ownership reinforce this structure which prioritizes human authority. The author of a work possesses the first copyright ownership according to Section 177 except for specific cases which include employment and commissioned photographs and portraits and engravings and films and government works and public undertakings8.
Section 199 requires copyright assignment to be documented in writing with signatures and description of assigned work and rights which must include duration and territorial extent details. The requirement holds special importance for NFTs because token transfer cannot establish copyright transfer
4 The Copyright Act 1957, section 13
5 The Copyright Act 1957, section 14
6 The Copyright Act 1957, section 2(d)
7 The Copyright Act 1957, section, 17
8 Anjali Sharma, ‘AI-Generated Works and Copyright Ownership: The Indian Perspective’ (2026) Indian
Journal of Law and Legal Research (IJLLR)
9 The Copyright Act 1957, section 19
without a written assignment that meets the necessary standards. II. Statutory Framework in India
Copyright protection exists for original literary and dramatic works and musical and artistic works together with cinematographic films and sound recordings according to Section 13. Section 14 establishes copyright as a set of exclusive rights that permit reproduction through any material medium and electronic storage and public distribution and public broadcasting and work adaptation and the electronic storage of artistic works and films and their public broadcasting. The regulations provide sufficient coverage to handle all activities which involve digital copying and online display and NFT-related distribution and machine-readable storage of artistic works. Section 2(d)10 functions as an essential element because it establishes different definitions of "author" for various types of works.
The author of literary or dramatic works contains all musical works which the composer created and all artistic works except photographs which the artist produced and all photographs which the photographer captured and all films and sound recordings which the producer created and all computer-generated works which the author created. This final phrase establishes a legal basis for assessing AI-generated creative works but it does not define the necessary amount of human supervision that needs to exist. Humanity-cantered systems exist because ownership regulations support them. The author becomes the initial copyright owner according to Section 1711 but specific exceptions apply to this rule which include works created during employment and commissioned photographs and portraits and engravings and films and government works and public projects.
Section 1912 requires copyright assignments to be documented through written agreements which must bear signatures and describe the assigned work and rights together with their duration and geographic scope. The requirement holds increased importance for NFTs because token transfer does not result in copyright transfer without a proper written assignment that meets the legal requirements.
10 The Copyright Act 1957, section 2(d)
11 The Copyright Act 1957, section 17
12 The Copyright Act 1957, section 19
III. NFTs and the Separation of Token Ownership from Copyright
The best understanding of an NFT exists through its definition as a blockchain-based token which connects to digital assets and their associated metadata but does not provide ownership rights to the original content. Indian legal experts explain that NFT transactions create confusion because customers believe they receive all rights to use the underlying artwork when they purchase the token. According to Indian copyright law the general public believes that assumption is true but copyright protection needs an official written transfer of rights which must match the requirements of Sections 1813 and 1914 of the law. The three common situations that require this distinction. The first situation allows a person who does not own copyright to create an NFT because he wants to mint an NFT for work.
The second situation exists because copyright law protects more than physical copies which means the owner who holds a legal copy cannot use his copy rights to create and sell digital assets. The third situation occurs because marketplace operators who sell infringing NFTs may face legal action for platform liability but they can protect themselves through section 79 intermediary protections of the Information Technology Act if they follow proper procedures and notify parties according to established rules. 15 From a doctrinal standpoint, NFTs should not be treated as a new species of copyright subject matter16. The better view is that they are digital instruments of authentication which serve as a means to prove scarcity while executing contracts for existing protected works. The legal analysis requires determining which actions occurred to the original work.
The questions should proceed through three testable stages which demonstrate all aspects of Sections 1417 and 19 without creating any technological exceptions.
IV. Generative AI and Authorship
The main legal challenge in India centres on determining the copyright status of output generated by generative AI systems. Section 2(d)(vi)18 grants authorship rights to whoever
13 The Copyright Act 1957, section 18
14 The Copyright Act 1957, section 19
15 Adnan Mujević and Amina Mujević, ‘NFTs and Copyright Law’ (2022)
16 ‘Tokenized Art: The Implications of Copyright Law on NFTs’ (2025) 30(4) Journal of Intellectual Property Rights (JIPR)
17 The Copyright Act 1957, section 18
18 The Copyright Act 1957, section 2(d)(i)
initiates the creation of computer-generated literary dramatic musical or artistic pieces. Indian law does not require direct manual fixation for every situation according to that statement. The system requires human involvement which has legal significance but it does not treat the AI system as an authorized creator who holds rights. To explain AI output three distinct categories should be established. The first category includes completely automatic output which results from minimal user input without the user having any control over how the content will be arranged or changed or presented. The second category includes AI-assisted output which requires human users to create prompts and define limits while making their selections and modifying the final product.
The third category includes commercial systems which involve multiple users who work together with developers and system operators; in these situations courts must identify the primary creative party who holds legal rights instead of giving copyright to software developers or platform operators. Section 17 structure supports this method. The Act first establishes that authors hold primary ownership until their employment or contract-based conditions transfer rights to others. The practice of designating AI creators as primary authors for all outputs would eliminate the boundary between creating tools and producing actual products. The camera manufacturer does not become the owner of all photographs taken with its equipment. V. Originality in AI-Inflected Works
Indian copyright law needs work because it requires originality but the existing requirements do not fulfil this requirement. Section 13 protects original literary, dramatic, musical and artistic works, and Indian doctrine has long treated originality as turning on human skill, labour and judgment rather than mere independent existence of the work. Originality in generative AI requires more than model output which produces statistically new results that do not match training data. The original doctrinal statement would define AI-assisted works as containing original content when a human creator used prompt design, curation, arrangement, editing, transformation, sequencing, or combination to create content that human judgment could trace. Simple prompts which demand creators to "create a sunset in watercolour style" do not meet actual requirements.
Multiple steps of layered prompting together with several rounds of testing and approval processes and total text modifications and human editing work together to show
creative decisions which meet the originality standard. The standard prevents machines from creating excessive patent rights which would overwhelm the copyright system with inferior patent rights. The method secures authentic creator AIsupported work while it rejects nonsensical outputs because they require minimal human effort which functions like pushing a button. The established threshold needs more emphasis in India because copyright law exists as a human-centric system while copyrighting nearly human complete work would create challenges for defining authorship and for establishing the legal reason of exclusive rights19. VI. Infringement Tests for NFTs and Generative AI
The section about digital rights violations for NFTs and generative AI needs to examine three different aspects which include three different steps of the process which begin with the actual materials and end with the final results which reach the public. The input stage asks whether the process of ingesting or scraping or storing or training people uses copyrighted content without permission to create digital copies of protected material according to Section 14. The Indian academic work about generative artificial intelligence shows that commercial training at large scale needs permission to operate according to the laws defined in the Act which protects digital reproduction and only provides specific exceptions to users instead of allowing unrestricted fair use access.
The output stage examines whether an AI system generates content which includes either complete or essential portions of copyrighted material. The courts should determine whether someone took protected expression or if artists shared their style with others through their work and then they need to assess how much alike the two works really are. A model output that merely reflects a genre, mood, or unprotectable style should not be treated as infringing merely because it resembles an artist’s “look. ” An output which generates material through probabilistic methods can still meet infringement requirements when it reproduces protected compositional elements and expressive arrangements and textual passages. At the dissemination stage, NFTs introduce a distinct but related inquiry.
NFT minting and listing processes need to store and display the original work which becomes accessible to the public so this requirement establishes public access which connects to copyright rights through the token because it holds original work. The main error 19 Eastern Book Company v D B Modak (n 15); ‘Balancing Indian Copyright Law with AI-Generated Content’ (n 17).
occurs when people treat the blockchain token as a separate legal entity which exists apart from the copyrighted material because actual market platforms use display pages and preview images and metadata links and promotional content to show exclusive rights to their users. VII. Re-examining the Infringement Standard
Indian courts should adapt existing infringement principles rather than inventing a sui generis AI exception. The three necessary improvements will create better results. The first requirement requires courts to recognize model training as a separate process from model output because both activities involve different types of evidence and rights protection. The second requirement demands that plaintiffs in AI-output cases present particular details about their protected expressive elements instead of making aesthetic infringement claims based on visual similarity. The third requirement states that NFT disputes should be treated as traditional copyright lawsuits which address unauthorized digital distribution. A person who mints a token linked to another’s protected artwork does not escape liability
merely because the token is unique; uniqueness of the token says nothing about licence to reproduce or communicate the underlying work. Platform operators should face liability based on their knowledge and ability to control systems and their efforts to investigate and deal with problems instead of their self-description as neutral blockchain intermediaries. VIII. Toward a Coherent Indian Approach
The Indian position requires four fundamental propositions to establish a unified framework. The first proposition states that AI systems function as tools which should receive legal authorship based on human creative work defined in Section 2(d)(vi). The second proposition states all AI-assisted works require evaluation of their originality through evidence of human evaluation instead of assessment through machine-generated results and basic text input. The third proposition states that NFT transfers do not automatically transfer copyright ownership because there must exist a written agreement that meets the requirements defined in Section
19. The existing bundle of Section 14 rights requires separate analysis of infringement activities
during training output generation and tokenized distribution processes. The propositions maintain doctrinal consistency while creating opportunities for upcoming changes in legislation. Parliament may eventually choose to introduce explicit text-and-data-
mining exceptions, AI transparency obligations, or platform-specific NFT compliance rules. The courts and scholars should maintain their current boundaries of copyright law until new technological advancements bring about legislative changes. The established legal framework requires us to apply its rules with complete transparency about technological progress.
IX. Conclusion
The Indian copyright law remains intact despite NFTs and generative AI because these technologies create uncertain legal interpretations through their capacity to identify authors and evaluate originality and track unauthorized reproduction. The statutory text already supplies a workable foundation: originality remains a threshold requirement under Section 13, the bundle of exclusive rights under Section 14 captures most relevant digital acts, Section 2(d)(vi) accommodates computer-generated works through a human-causation model, and Sections 18-
19 prevent informal assumptions that token sales transfer copyright.
BIBLIOGRAPHY
Journal articles and working papers • ‘PROTECTING COPYRIGHT IN DIGITAL ART AND NFTS’ (2024) International Journal of Creative Research and Technology (IJCRT) • ‘Tokenized Art: The Implications of Copyright Law on NFTs’ (2025) 30(4) Journal of Intellectual Property Rights (JIPR) • Adnan Mujević and Amina Mujević, ‘NFTs and Copyright Law’ (2022) (preprint) • Anjali Sharma, ‘AI‐Generated Works and Copyright Ownership: The Indian Perspective’ (2026) Indian Journal of Law and Legal Research (IJLLR) • ‘NFTs and Copyright: Legal Challenges in Virtual Assets’ (2025) Indian Journal of Legal Research (IJLR) • ‘AI‐Generated Works and Copyright in India’ (2026) (working paper) • ‘Copyright Issues of AI‐Generated Works’ (2026) (student paper) • Nivedita Ramachandran, ‘Navigating New Frontiers: The Impact of NFTs on Copyright Law in the Digital Era’ (2023) MHU Journal of Innovation and Research Blogs, law‐firm notes and online commentary • ‘Balancing Indian Copyright Law with AI‐Generated Content: The Significant Human Input Approach’ (NLSIU IJLT Blog, 7 August 2025) • ‘NFTs and Intellectual Property: Legal Protection of Digital Art in India’ (Khurana & Khurana, 4 January 2026) • ‘Ownership of AI Generated Content: A Deep Dive into Copyright Law in India’ (Khurana & Khurana, 24 October 2025) • ‘AI‐Generated Content & Copyright Law in India’ (IPLink Asia, 4 April 2025)
‘Ownership and Authorship of AI‐Generated Works under Indian Copyright Law: A Regulatory Gap’ (Vintage Legal, 21 March 2026) Legislations Copyright Act 1957 Information Technology Act, 2000