AI-Generated Works And The Challenge Of Copyright Ownership

Author: LawSphereX

Category: Intellectual Property

Published:

Updated:

AI-GENERATED WORKS AND THE

CHALLENGE

OF COPYRIGHT OWNERSHIP

U. Jeevitha, Advocate in Tirunelveli District Court

ABSTRACT

The rapid advancement of artificial intelligence (AI) has transformed the creative landscape, enabling machines to generate music, literature, art, and other forms of expressive works that traditionally relied on human authorship. While these developments present unprecedented opportunities for innovation and efficiency, they also challenge the very foundations of copyright law. Conventional copyright systems are built upon principles of human creativity, originality, and authorship, making it unclear whether AIgenerated works fall within existing legal frameworks. This paper examines the legal, ethical, and economic complexities surrounding the copyright status of AI-created content, with particular attention to global legislative approaches and judicial interpretations.

Through a comparative analysis of jurisdictions including the United States, European Union, United Kingdom, India, and selected Asia-Pacific nations, the study explores how different legal systems address questions of authorship, ownership, originality, and liability in the context of AI-generated works. It further analyzes the implications for creative industries, human artists, and public policy, highlighting the tension between protecting human creativity and fostering technological innovation. The paper also evaluates doctrinal perspectives such as the labour theory of property, utilitarian justifications, and the role of the public domain in an AI-driven creative environment. Drawing on global best practices, it proposes potential reforms, including hybrid authorship models, international harmonization of copyright laws, and ethical guidelines for AI use in creative fields.

Ultimately, the study argues that while AI challenges traditional copyright principles, it also offers an opportunity to modernize legal frameworks to reflect emerging technological realities. Keywords: Artificial Intelligence, Copyright Ownership, AI-Generated Works, Authorship, Intellectual Property Law

INTRODUCTION

The emergence of artificial intelligence (AI) as a creative tool has redefined the boundaries of artistic and intellectual expression, challenging long-standing assumptions within copyright law. From generating complex musical compositions and realistic paintings to producing literature and film scripts, AI systems now possess the capacity to produce works that are indistinguishable from those created by humans. However, this technological evolution raises fundamental questions regarding the applicability of existing copyright principles, which have historically been premised on human authorship, originality, and the expression of personal creativity.

The absence of human intervention in certain AI-generated works prompts debates about whether such creations can be afforded legal protection at all, and if so, who should rightfully hold the ownership—whether it be the AI’s programmer, the user operating the system, or the public domain. Moreover, the global legal landscape is fragmented, with jurisdictions adopting divergent approaches to defining authorship and determining the scope of protection for AI-generated content. These differences not only create uncertainty for creators, technology developers, and industries but also hinder the development of consistent international norms.

In addition to legal challenges, the proliferation of AI in creative fields has profound ethical and economic implications, including the potential displacement of human artists, concerns over cultural authenticity, and the risk of monopolization by technology-driven entities. Against this backdrop, the present study seeks to critically analyze the complexities surrounding copyright ownership in AI-generated works, examining comparative legal frameworks, exploring doctrinal theories, and assessing policy considerations. The paper also aims to propose viable reforms that balance the promotion of technological innovation with the preservation of human creativity, while advocating for harmonized global standards that can effectively address the evolving realities of the digital age.

UNDERSTANDING AI AND COPYRIGHT Defining Artificial Intelligence in the Creative Sphere Artificial Intelligence (AI) in the creative sphere refers to computer systems and algorithms designed to mimic human cognitive abilities in producing creative outputs such as music, literature, visual art, and design. Unlike traditional software that follows pre-defined instructions, AI systems—especially those using machine learning and deep neural networks— can analyze large datasets, identify patterns, and generate original works that may not be explicitly programmed. In the Indian context, AI creativity is gaining recognition in industries such as advertising, film, and digital art, where platforms like generative adversarial networks (GANs) and natural language processing (NLP) tools are used to create novel content.

However, the Indian Copyright Act, 1957, does not explicitly define AI or recognize nonhuman creators, which creates a legislative gap in determining authorship. The closest statutory reference is Section 2(d)1 of the Act, which defines “author” but confines it to human entities, raising questions about how AI-generated works fit within this framework. Types of AI-Generated Works (Fully AI-Created vs. AI-Assisted) AI-generated works can broadly be divided into two categories: fully AI-created works and AI-assisted works. Fully AI-created works are generated without substantial human input beyond initial programming or data training—such as an AI composing a symphony independently. In contrast, AI-assisted works involve human creativity at significant stages, where AI is merely a tool, for example, a designer using AI-powered software to refine a visual concept.

Indian law currently treats AI-assisted works as human-authored since there is identifiable human input. The case Eastern Book Company v. D. B. Modak,2 indirectly relates here by stressing that originality requires a “modicum of creativity” from the author. While this case did not involve AI, its reasoning suggests that purely autonomous AI outputs may struggle to qualify for copyright unless human involvement in creative decision-making is evident. Traditional Copyright Principles – Authorship, Originality, and Fixation Copyright law rests on three pillars—authorship, originality, and fixation. In India, Section 133 of the Copyright Act grants protection only to original works of authorship that are fixed in a tangible medium. The concept of authorship is defined in Section 2(d), which presupposes a natural person as the creator.

Originality, as interpreted in University of London Press Ltd. v. University Tutorial Press Ltd. ,4, requires independent creation and minimal creativity—a standard echoed in Indian jurisprudence. The Eastern Book Company case

1 Section 2(d) of the Copyright Act https://indiankanoon.org/doc/1136195/

2 Eastern Book Company v. D.B. Modak, (2008) 1 SCC 1

3 Section 13 of the Copyright Act https://indiankanoon.org/doc/1136195/

4 University of London Press Ltd. v. University Tutorial Press Ltd., (1916) 2 Ch 601

further refined originality in India, emphasizing creativity rather than mere labor. Fixation is satisfied when the work is recorded in a perceivable form, which AI outputs can easily achieve. However, if an AI independently generates a fixed work without human creativity, it challenges the principle that authorship must stem from a human mind—a gap Indian law has yet to address. Intersection of AI Technology and Copyright Law The intersection of AI technology and copyright law is a rapidly evolving legal frontier in India. AI tools can replicate styles, generate derivative works, or create entirely novel expressions, raising potential infringement and ownership issues. For instance, if an AIgenerated song resembles an existing copyrighted melody, determining liability becomes complex—should it lie with the programmer, the user, or neither?

While Indian courts have not yet adjudicated a direct AI authorship case, principles from related decisions provide guidance. In Indian Performing Right Society Ltd. v. Eastern Indian Motion Pictures Association5, the Supreme Court recognized the economic rights of authors and emphasized that these rights cannot exist without a creator. This reasoning indirectly suggests that AI outputs without a human creator may fall outside the current copyright scope. Moreover, Section 176 of the Copyright Act, which deals with ownership in works created under employment or commission, could potentially be extended to contractual arrangements involving AI, where the commissioning party might claim rights. However, absent explicit legislative recognition, the legal position remains uncertain, highlighting the urgent need for policy reforms to address AI’s role in creative industries.

GLOBAL

LEGAL FRAMEWORK

ON AI-GENERATED WORKS The global legal framework on AI-generated works examines how different jurisdictions treat the copyright status of creative outputs produced by artificial intelligence. While some countries strictly require human authorship for copyright protection—such as the United States, European Union, Australia, Japan, and India—others have adopted more flexible approaches. The United Kingdom, through Section 9(3)7 of its Copyright, Designs and Patents Act 1988, attributes authorship of computer-generated works to the person making 5 Indian Performing Right Society Ltd. v. Eastern Indian Motion Pictures Association, (1977) 2 SCC 820

6 Section 17 of the Copyright Act https://indiankanoon.org/doc/1136195/

7 Section 9(3) of the Copyright Act https://indiankanoon.org/doc/1136195/

the necessary arrangements. China has even recognized AI-generated works in certain cases, such as the Shenzhen Tencent v. Shanghai Yingxun8 decision. These differences create challenges for cross-border enforcement and harmonization of laws. While human-centric systems aim to preserve traditional notions of creativity, more inclusive approaches seek to adapt copyright law to technological realities. The divergence highlights the urgent need for international dialogue and policy reform to address authorship, originality, and ownership in the age of AI. Position in the United States In the United States, copyright protection is governed by the Copyright Act of 1976, which requires “original works of authorship” fixed in a tangible medium. The U. S. Copyright Office (USCO) has made it clear that works created solely by AI without human authorship are not eligible for protection.

This was reaffirmed in Thaler v. Perlmutter (2023)9, where the U. S. District Court held that a work generated by Stephen Thaler’s “Creativity Machine” could not be copyrighted because the statute requires human creativity. However, AI-assisted works may still qualify if there is sufficient human selection, arrangement, or modification in the creative process. The USCO’s Compendium of Copyright Office Practices (Third Edition) also provides that AI-generated material lacking human intervention does not meet the originality standard. This human-centric approach aims to protect human creators while avoiding ambiguity over machine-generated authorship. Approach in the European Union The European Union follows the Copyright Directive (2019/790) and related jurisprudence of the Court of Justice of the European Union (CJEU).

The EU requires “the author’s own intellectual creation,” as established in Infopaq International A/S v. Danske Dagblades Forening10, meaning the work must reflect the personality and creative choices of a human author. This standard effectively excludes fully autonomous AI works from protection, though AI-assisted works can qualify if human creative input is evident. The EU is also moving forward with the Artificial Intelligence Act, which, while not directly granting copyright,

8 Shenzhen Tencent v. Shanghai Yingxun

9 Thaler v. Perlmutter (2023)

10 Infopaq International A/S v. Danske Dagblades Forening (C-5/08)

seeks to regulate AI transparency, risk assessment, and accountability—indirectly influencing copyright disputes involving AI outputs. Developments in the United Kingdom The United Kingdom’s position is unique due to Section 9(3) of the UK Copyright11, Designs and Patents Act 1988 (CDPA), which explicitly states that for “computer-generated works,” the author is deemed to be “the person by whom the arrangements necessary for the creation of the work are undertaken. ” This provision potentially allows AI-generated works to be protected, with ownership attributed to the individual or entity controlling the creative process. However, the UK Intellectual Property Office (UKIPO) has launched consultations on whether this provision adequately addresses modern AI capabilities.

While no landmark court decision has directly interpreted Section 9(3) in an AI-only context, it provides a legislative basis for recognizing non-human-assisted outputs, setting the UK apart from the EU and US. Indian Legal Position In India, the Copyright Act, 1957 does not contain any specific provisions for AIgenerated works. Authorship under Section 2(d) presupposes human agency, and originality under Section 13 is tied to human creativity. Although India has not yet faced a direct judicial determination on AI authorship, related reasoning can be drawn from R. G. Anand v. Delux Films12, where the Supreme Court emphasized that copyright exists only in the expression of an idea and requires an identifiable author. Given this framework, fully autonomous AI creations would likely not be protected unless there is demonstrable human contribution.

Moreover, the increasing role of AI in Bollywood music, visual effects, and design industries has triggered debates among policymakers about amending the law to include AI-generated works, possibly by adopting a UK-style approach where the commissioning party or AI operator could be deemed the author. Comparative Insights – Australia, Japan, and China Australia applies the Copyright Act 1968, which, according to cases like Telstra

11 Section 9(3) of the UK Copyright https://www.legislation.gov.uk/ukpga/1988/48/contents

12 R.G. Anand v. Delux Films, (1978) 4 SCC 118

Corporation Ltd v. Phone Directories Co Pty Ltd (2010)13, requires human authorship. This effectively excludes autonomous AI creations, though AI-assisted works may be eligible if human input is creative. Japan adopts a more technology-neutral stance: while copyright requires human authorship, the Japanese Copyright Act (2019 amendment) introduced special rules for the use of data in AI training, encouraging innovation without breaching existing rights. China has taken a more proactive approach—in Shenzhen Tencent v. Shanghai Yingxun (2019)14, a Chinese court recognized copyright in an article generated by Tencent’s AI “Dreamwriter,” reasoning that it involved intellectual investment and met originality standards, although this decision is not uniformly applied across all jurisdictions in China.

These variations show that while some countries like the UK and China lean towards accommodating AI-generated works, others like the US, EU, and India maintain a stricter human-authorship requirement. Comparative Table: Global Legal Approaches to AI-Generated Works JURISDICTION HUMAN AUTHORSHIP REQUIRED?

PROTECTION FOR

FULLY AI-CREATED WORKS LEGAL BASIS / CASE REFERENCE UNITED STATES Yes No Thaler v. Perlmutter (2023); Copyright Act 1976 JAPAN Yes No (but AI-friendly data use rules) Japanese Copyright Act (2019 amendment) CHINA Not strictly Yes (case-specific) Shenzhen Tencent v. Shanghai Yingxun (2019)

CORE LEGAL

CHALLENGES

Defining “Authorship” in the Age of AI The concept of authorship is central to copyright law, yet AI challenges its very foundation. Under Section 2(d) of the Indian Copyright Act, 1957, an “author” is defined according to the category of the work, always referring to a natural or legal person, never a machine. AI-generated works without significant human intervention thus face a legal vacuum. In Civic Chandran v. Ammini Amma, 1996 PTC15, the Kerala High Court stressed the need for identifying a creator whose intellectual effort is embedded in the work. Applying this reasoning, an AI system cannot independently be considered an “author” because it lacks human intellect and legal personality. This uncertainty makes authorship in AI contexts a contested legal territory.

Originality and Human Creativity Requirement Originality requires an element of intellectual creativity, beyond mere labor or skill. In India, this is a well-established principle reinforced in Macmillan & Co. Ltd. v. K. & J. Cooper16, where the Privy Council held that originality requires a work to originate from the author and not be a mere copy. For AI-generated works, the absence of human creativity challenges this requirement. While AI may produce novel content, its originality is algorithmic, derived from training data and patterns, raising questions about whether it truly satisfies the “intellectual creation” test under Indian law. Ownership Disputes – Developer vs. User vs. Public Domain When AI produces a work, disputes arise as to whether ownership belongs to the

15 Civic Chandran v. Ammini Amma, 1996 PTC (16) 329 (Ker)

16 Macmillan & Co. Ltd. v. K. & J. Cooper, AIR 1924 PC 75

AI’s developer, the end-user, or whether the work should enter the public domain. Section 17 of the Copyright Act outlines ownership rules in employer-employee and commissioned works but does not address AI scenarios. In Najma Heptulla v. Orient Longman Ltd. 17, the Delhi High Court addressed contractual allocation of rights between contributors and publishers, underscoring the importance of clear agreements. Similarly, AI-generated works may require pre-defined contractual terms to avoid disputes, particularly in collaborative projects involving multiple stakeholders. Infringement and Liability Issues AI can inadvertently create works that resemble existing copyrighted material, leading to infringement claims. Determining liability is complex—should it rest with the programmer, the operator, or both?

Indian law, under Sections 51 and 55 of the Copyright Act18, imposes liability on persons who commit or authorize infringement. In Super Cassettes Industries Ltd. v. Hamar Television Network Pvt. Ltd. ,19, the Delhi High Court stressed the principle of authorization in infringement, suggesting that if AI output violates copyright, the party controlling or authorizing its operation could be liable. This becomes even more complex with generative AI, where outputs are created autonomously from vast datasets. Moral Rights in AI-Generated Works Moral rights, protected under Section 57 of the Copyright Act20, safeguard an author’s right to claim authorship and to protect the integrity of the work. In Amarnath Sehgal v.

Union of India21, the Delhi High Court upheld the moral rights of a sculptor whose artwork was dismantled without consent, emphasizing the personal connection between the author and the work. Since AI lacks personal identity or feelings, it cannot possess moral rights; however, disputes may arise when human collaborators wish to assert such rights over AI-assisted works. This could lead to hybrid moral rights claims where human creators assert authorship over collaborative outputs involving AI systems.

17 Najma Heptulla v. Orient Longman Ltd., AIR 1989 Del 63

18 Sections 51 and 55 of the Copyright Act https://indiankanoon.org/doc/1136195/

19 In Super Cassettes Industries Ltd. v. Hamar Television Network Pvt. Ltd., 2011 (47) PTC 49 (Del)

20 Section 57 of the Copyright Act https://indiankanoon.org/doc/1136195/

21 Amarnath Sehgal v. Union of India, 2005 (30) PTC 253 (Del)

DOCTRINAL AND THEORETICAL PERSPECTIVES Labour Theory of Property The Labour Theory, articulated by John Locke, holds that property rights arise when a person mixes their labor with resources from the commons. In the copyright context, this means the creator earns ownership by investing skill, effort, and time into producing a work. Under Section 1322 of the Indian Copyright Act, originality and authorship are rooted in human contribution. In V. T. Thomas v. Malayala Manorama Co. Ltd23, the Kerala High Court recognized the creative labor of a cartoonist as the source of his ownership rights. However, for AI-generated works, where human labor is minimal or limited to initiating the process, the Lockean justification becomes strained. If no substantial human effort is evident, the work may default to the public domain.

Utilitarian Justification of Copyright The utilitarian theory views copyright as a tool to promote public welfare by incentivizing creation through limited monopolies. The rationale is codified in Article 7 of the TRIPS Agreement, which India adheres to, emphasizing that IP rights should promote technological innovation and social benefit. In Entertainment Network (India) Ltd. v. Super Cassette Industries Ltd. 24, the Supreme Court balanced exclusive rights with public interest, allowing compulsory licensing in certain cases. Applying this to AI, granting copyright could incentivize AI development, but excessive monopolization could stifle innovation and restrict access to knowledge. Personality Theory and AI-Created Content The Personality Theory, associated with Hegel, argues that creative works reflect the author’s personality and moral identity.

Indian law recognizes this connection through Section 5725 (moral rights). In Kamalapati Trivedi v. State of West Bengal26, the court protected the moral rights of a playwright, acknowledging the deep personal link between the creator and

22 Section 13 https://indiankanoon.org/doc/1136195/

23 V.T. Thomas v. Malayala Manorama Co. Ltd., 1988 AIR 37 (Ker)

24 Entertainment Network (India) Ltd. v. Super Cassette Industries Ltd., (2008) 13 SCC 30

25 Section 57 https://indiankanoon.org/doc/1136195/

26 Kamalapati Trivedi v. State of West Bengal, AIR 1980 Cal 98

the work. Since AI lacks personality, emotions, and self-expression, this theory cannot directly justify granting it authorship. However, it can still support the rights of human-AI collaborators, where human creativity shapes the AI’s output. Public Domain and Access to Knowledge The public domain ensures that works not protected by copyright remain freely accessible for education, research, and cultural growth. Section 52 of the Copyright Act also allows certain fair use exceptions that promote access to knowledge. In Academy of General Education v. B. Malini Mallya27, the Supreme Court emphasized balancing private rights with public interest. AI-generated works without human authorship could default to the public domain, ensuring free availability while avoiding monopolization by AI developers.

However, this raises concerns about commercial exploitation without rewarding the parties who facilitated AI creation. Chart – Doctrinal Application to AI-Generated Works THEORY CORE IDEA APPLICATION TO AI WORKS Labour Theory Ownership arises from effort and skill Weak in fully autonomous AI outputs

PROPOSED SOLUTIONS AND

RECOMMENDATIONS

Proposed solutions to address the legal complexities surrounding AI-generated works require a balanced approach that integrates legislative reforms, hybrid authorship models, international cooperation, and ethical guidelines. Legislative reforms can draw from global best practices, such as the European Union’s proposed AI Act and the UK Intellectual Property Office’s guidelines on computer-generated works, while tailoring them to India’s legal framework under the Copyright Act, 1957. This may involve explicitly defining AI-generated works, clarifying authorship criteria, and setting liability standards.

A hybrid model of authorship and joint ownership—where both the human contributor (such as the programmer, trainer, or curator) and the AI system’s deploying entity share rights—can address disputes over ownership, especially in cases where significant human input coexists with autonomous AI creativity. International harmonization is equally crucial, as cross-border dissemination of AI content raises jurisdictional conflicts; aligning India’s policies with instruments like the Berne Convention and TRIPS Agreement would promote consistency and mutual recognition of rights. Ethical guidelines should emphasize transparency, accountability, and respect for existing rights, including moral rights under Sections 57 of the Copyright Act.

For instance, adopting a framework akin to the OECD Principles on AI could ensure that AI tools in creative sectors are deployed responsibly, with safeguards against plagiarism, cultural misappropriation, and biased outputs. Indian jurisprudence, while limited on AI-specific issues, provides analogies—such as Eastern Book Company v. D. B. Modak28, which recognized the importance of skill and judgment in determining originality—that could guide policy in differentiating between purely machine-generated content and human-assisted creation. Ultimately, these reforms and frameworks should aim to foster innovation while protecting creators’ rights, ensuring AI remains a tool for augmenting human creativity rather than replacing or undermining it.

CONCLUSION

In the emergence of AI-generated works presents both unprecedented opportunities and complex legal challenges that demand urgent attention from lawmakers, industry stakeholders,

28 Eastern Book Company v. D.B. Modak (2008) 1 SCC 1

and the global community. While AI’s capacity to create original-like content has expanded the boundaries of human imagination, it has also disrupted traditional concepts of authorship, originality, and ownership that underpin intellectual property regimes. The existing legal frameworks, including the Indian Copyright Act, 1957, were primarily designed for human creators, leading to ambiguities when applied to non-human outputs. Global best practices and doctrinal perspectives suggest that a balanced approach—recognizing the role of human input, ensuring fair attribution, and protecting the public domain—is essential to maintain both innovation and equity. Harmonizing international standards, introducing hybrid authorship models, and implementing ethical AI guidelines will be crucial to prevent exploitation and ensure accountability.

Ultimately, the legal treatment of AI-generated works must strive to safeguard creators’ rights, promote technological progress, and preserve public access to knowledge. Without proactive reforms, the risk of legal uncertainty, monopolization, and cultural erosion may overshadow the transformative potential of AI in creative domains. This is a pivotal moment for shaping a future where human creativity and machine intelligence can coexist symbiotically within a fair and sustainable legal framework.

REFERENCE:

Bibliography

1. Narayanan, P. (2021). Intellectual Property Law. Eastern Book Company, Lucknow.

2 Cornish, W.R., Llewelyn, D., & Aplin, T. (2019). Intellectual Property

Patents, Copyright, Trade Marks and Allied Rights (9th ed.). Sweet & Maxwell.

3. Basheer, S., & Reddy, P. (2020). Innovation, Intellectual Property and Development: A

Legal and Economic Analysis. Oxford University Press, New Delhi.

4 Jain, A. (2022). Copyright Law in India

Concepts and Cases. Universal Law Publishing. Webliography

1. World Intellectual Property Organization (WIPO). (2024). “Artificial Intelligence and

Intellectual Property.” Available at: https://www.wipo.int

2. Department for Promotion of Industry and Internal Trade (DPIIT), Ministry of

Commerce and Industry, Government of India. “Copyright Office – Official Website.” Available at: https://copyright.gov.in

3. European Parliament Research Service. (2022). “Intellectual Property Rights and

Artificial Intelligence.” Available at: https://www.europarl.europa.eu Books

1. Scardamaglia, A., & Fitzgerald, B. (2022). The Future of Intellectual Property Law and

Artificial Intelligence. Edward Elgar Publishing.

2 Kur, A., & Dreier, T. (2013). European Intellectual Property Law

Text, Cases and Materials. Edward Elgar Publishing. Articles

1. Ginsburg, J.C., & Budiardjo, L. (2019). “Authors and Machines.” Columbia Journal of

Law & the Arts, 42(4), 615–650.

2. Samuelson, P. (2022). “Copyright and Creativity in the Age of Artificial Intelligence.”

Journal of the Copyright Society of the USA, 69(3), 231–272.

3. Menell, P.S. (2021). “Governance of Intellectual Property Rights in the Era of AI.”

Berkeley Technology Law Journal, 36(1), 1–50. Journals

1. Journal of Intellectual Property Rights (CSIR-NIScPR, India).

2. Indian Journal of Law and Technology (National Law School of India University).

3. Harvard Journal of Law & Technology.