Decoding The Ownership Dilemma - Copyrights And AI-Generated Artworks In The Metaverse

Author: LawSphereX

Category: Intellectual Property

Published: 2026-08-23T19:12:52.322Z

Updated: 2026-08-30T07:12:52.727Z

DECODING THE OWNERSHIP DILEMMA: COPYRIGHTS

AND AI-GENERATED ARTWORKS IN THE METAVERSE

Rushil Verma, Symbiosis Law School, Pune

ABSTRACT

This article explores how Artificial Intelligence (AI) and the Metaverse are reshaping the way the we think about creativity and copyright. It begins by explaining that AI can now produce art, music, and writing on its own, while the Metaverse provides immersive virtual spaces where these works can be shared, bought, and sold. Traditional laws, like the Indian Copyright Act,

1957 and international treaties, assume only humans can create original

works. This raises questions: who owns AI-generated art, and are such creations even protected? Next, the paper examines real-world examples and court cases – from India, the U.S., and the U.K. – that highlight gaps in authorship rules and enforcement challenges in decentralized virtual worlds. It also discusses ethical issues, such as bias in AI training data and the “black‐box” nature of algorithms. Finally, the paper proposes solutions: redefining authorship to include human “AI‐assisted” creators, using blockchain and smart contracts for real‐time rights management, and creating an international treaty to harmonize AI copyright rules. Emphasizing fairness and transparency, the article calls for urgent legal reforms to balance innovation with respect for creators.

INTRODUCTION

The intersection of Artificial Intelligence (AI) and the Metaverse represents a pivotal evolution in how societies view creativity, authorship, and intellectual property. AI, the intelligence exhibited by machines, at present is generative enough to go beyond traditional functions of computation to become a producer of art, literature, music, and other forms of intellectual property. Furthermore, Metaverse is a network of inter-connected digital worlds meant for users to enjoy immersive experiences and virtual interactions, which are almost real-life-like. Intellectual properties can be exhibited, purchased, sold, and enjoyed here. There are many ongoing developments in the Metaverse, which has raised crucial doubts about the traditional Intellectual Property Rights (IPR), making it imperative to re-evaluate and update the current laws in place.

The Copyright Act, 19571 as well as international treaties like the Berne Convention2 have traditionally presumed that unwritten IP work can only come into existence through human intellect which includes human authorship, human creativity, and human origin, where conscious thought comprises skills and originality towards these activities. However, the assertions mentioned above have been challenged by AI technologies since they are currently generating works produced independently or partially independently by machines. Thus, it becomes a pertinent question in this context as to who would own the copyright for a digital artwork generated by an AI system that is trained on large sets of existing works? Does the ownership belong to the AI programmer or the user of the AI, or does it come under the exception of not being protected by the existing laws?

Such challenges are further caste into darkness in the Metaverse, where jurisdictional boundaries fade and virtual assets gain realworld value. Currently, the Metaverse and AI industries represent massive potential economic importance, and the urgency of the issues under discussion is emphasized by this alone. Reports suggest that the Metaverse will provide trillions of dollars in economic activity within the next few decades and that AI technologies will be a key enabler of this growth. 3 As Pamela Samuelson argues in her article "Allocating Ownership Rights for AI-Generated Content," current

1 Copyright Act, 1957, No. 14, Acts of Parliament, 1957 (India).

2 Berne Convention for the Protection of Literary and Artistic Works (1886), 828 U.N.T.S. 221.

3 Carlos Cantú, Cecilia Franco, and Jon Frost, The Economic Implications of Services in the Metaverse, BIS Papers No. 144, 4 (2024). copyright law is, in many respects, inadequate and creates a cloud of uncertainty potentially stifling innovation and investment in the AI ecosystem. 4 With decentralized and interoperable architecture, the Metaverse exacerbates the challenges of enforcing intellectual property rights, while traditional enforcement mechanisms fail to apply readily for digital, transnational situations. The essay is going to pursue some of the deep challenges AI and the Metaverse present to copyright law. It will try to locate doctrinal gaps, assess the pros and cons of landmark case laws such as Feist Publications, Inc. v. Rural Telephone Service Co. 5 and Thaler v.

Comptroller-General of Patents, Designs, and Trademarks6, and consider literature works such as Ryan Abbott's "The Reasonable Robot: Artificial Intelligence and the Law". 7 This analysis aims to pinpoint concrete solutions that create a balance for strong intellectual property protection and for promoting innovation. Through a thorough discussion of authorship questions and enforcement issues including ethical concerns, the essay will act as a complete roadmap to reconciling the transformative ability of AI and the Metaverse with the law of intellectual property. The proposals for the respective possible solutions suggest a model for cooperation between legal reform, technological advancement, and international collaboration, fully attuned to the level of complexity and global character of the issues at hand.

THE LEGAL CONUNDRUM OF AUTHORSHIP IN AI-GENERATED CONTENT The historic premise of copyright law has always been authorship, which postulates that recognition and exclusive rights in respect of intellectual output should follow authors. But now, with the emergence of more popular acceptance, AI-related contents are breaking these grounds as it is increasingly accepted that AI systems can create literary, artistic, and music works just as well as humans. Section 2(d) of the Copyright Act, 19578, defines the term "author" as a person who makes a work, thereby indicating an inference of human creativity in this context. For this same reason, it almost bears a parallel statutory structure to that of the 4 Pamela Samuelson, Allocating Ownership Rights for AI-Generated Content, 47 U. Pitt. L. Rev. 1185, 1196 (1986).

5 Feist Publications, Inc. v. Rural Telephone Service Co., 499 U.S. 340 (1991).

6 Thaler v. Comptroller-General of Patents, Designs, and Trademarks, 2023 UKSC 49.

7 Ryan Abbott, The Reasonable Robot

Artificial Intelligence and the Law (Cambridge University Press 2020).

8 Copyright Act, 1957, § 2(d), No. 14, Acts of Parliament, 1957 (India).

Berne Convention9, in that neither anticipates nor accommodates creators that may not be human. A significant problem now arises when considering whether AI-generated works can be viewed as original. As the Supreme Court of the United States, in Feist Publications, Inc. v. Rural Telephone Service Co. 10, held, originality is independent creation combined with a small degree of creativity. So arises the question, if an AI system autonomously generates content without human intervention, can one say there exists an "original author"? The U. K. case of Thaler v. Comptroller General of Patents, Designs and Trademarks11 undertook a similar inquiry in the context of patents. In its judgment, the Supreme Court upheld the Comptroller’s decision declaring that an AI could not be regarded as an inventor since patent law requires a "natural person" as the originator of an invention.

This decision further augments the general legal unwillingness to bestow authorship or ownership rights to AI. Although AI authorship has yet to receive an unequivocal ruling in India, the Supreme Court has held, in the Eastern Book Company v. D. B. Modak12 case, that copyright requires a combination of human skill and acumen as opposed to mere mechanical effort. Therefore, the works produced by AI, which have no intervention by a human being, may be excluded from copyright protection under the Indian laws. Academicians have proposed diametrically opposing options to break the deadlock. Ryan Abbott, in his book entitled “The Reasonable Robot: Artificial Intelligence and the Law,” defended copyright registration for works created by AIs, with rights vesting in the person or corporate entity controlling or programming the AI.

13 Furthermore, Pamela Samuelson wrote in the article "Allocating Ownership Rights for AI-Generated Content" that while AI is not to be purported as an author, it is necessary to build a sui generis structure to allocate AI-generated works a fraction of rights to guarantee the remaining economic incentives. 14

9 Berne Convention, supra note 2.

10 Feist Publications, supra note 5.

11 Thaler, supra note 6.

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

13 Ryan Abbott, The Reasonable Robot

Artificial Intelligence and the Law 71-91 (Cambridge University Press 2020). 14 Pamela Samuelson, Allocating Ownership Rights for AI-Generated Content, 47 U. Pitt. L. Rev. 1185, 1187-

1190 (1986).

Without legal recognition for AI authorship, practical consequences ensue. Corporations that invest heavily in AI content cannot feel secure about their rights over it. The very lack of clarity leads to lack of investments and risks regarding enforcement and monetization within digital ecosystems like the Metaverse, where AI-generated assets emerge rapidly. European Parliament members have recognized this problem and a 2020 report on Intellectual Property Rights for the Development of Artificial Intelligence Technologies suggested a new rights framework that grants authorship to "AI-assisted" human creators rather than to AI itself. 15 A significant proposal for settling such matters relates to copyright management through blockchain.

Primavera De Filippi and Aaron Wright elaborate on this view in “Blockchain and the Law: The Rule of Code”, discussing how decentralized ledger technology permits the registering of AI-generated works with the attribution of authorship and ownership in a dynamic way based upon predefined governance rules. 16 This way, AI-generated works could be legally recognized without direct authorship assignment to AI itself, thus being compatible with the extant legal principles and accommodating technological changes. As AI gets more sophisticated and is used in deeper integration into content creation processes, it is imperative that legislators act soon.

Pro-active measures must be taken by policymakers so that the failure of copyright laws in adequately covering AI-assisted works, be it through amendments proposing the such, or new acknowledgment of a sui generis framework for AIgenerated content, will not provide an entry for actors into halt innovation. Harmonizing copyright laws worldwide with AI provisions in international treaties such as TRIPS17 will be critical in ensuring an internationally consistent treatment of AI-generated works. The doctrinal and practical challenges of AI-generated authorship necessitate a rethinking of copyright frameworks.

While courts and legislatures have yet to fully resolve these issues, scholarly discourse and international policymaking efforts suggest that a hybrid approach— one that assigns rights to human operators while acknowledging AI's role—may offer the most pragmatic solution to the authorship dilemma in the digital age. 15 Report on Intellectual Property Rights for the Development of Artificial Intelligence Technologies, European Parliament (2020), Motion for a European Parliament Resolution, ¶ J, 15. 16 Aaron Wright and Primavera De Filippi, Blockchain and the Law: The Rule of Code 179 (Harvard University Press 2019).

17 Agreement on Trade-Related Aspects of Intellectual Property Rights (1994).

CHALLENGES

IN THE METAVERSE The decentralized global aspect of the Metaverse does indeed complicate the enforcement of copyright laws. The traditional copyright frameworks apply to incidences of physical infringing acts or easily traceable digital acts. However, given the dynamic and anonymous flow of content in the Metaverse, there are hardly any avenues left for functional enforcement. Jurisdictional uncertainty tops the list of concerns: with multiple countries witnessing users, servers, and transactions, it becomes highly convoluted to determine the applicable law and relevant agencies responsible for its enforcement. 18 (Samuelson, Allocating Ownership Rights for AI-Generated Content). A prime example of these challenges is the unauthorized reproduction and distribution of AIgenerated artworks within the Metaverse.

Section 51 of the Copyright Act, 195719, prohibits unauthorized reproduction, adaptation, and communication of copyrighted works, but these provisions are difficult to apply in virtual environments. For instance, in Indian Performing Right Society v. Eastern India Motion Pictures Association20, the Supreme Court of India underscored the importance of enforceable copyright mechanisms. However, traditional copyright enforcement tools such as cease-and-desist notices and Digital Millennium Copyright Act, 199821 takedown requests prove inadequate against the decentralized nature of blockchain-based virtual worlds. 22 In addition, introducing non-fungible tokens (NFTs) within the metaverse has further entangled copyright enforcement. NFTs authenticate digital assets that assert ownership; copyright is not one of those guaranteed rights.

This has led to a cascade of litigation regarding art related to NFTs. The Hermès International v. Rothschild23 case is an example of a scenario in which NFTs may infringe trademark and copyright laws in the virtual environment. Likewise, Miramax v. Tarantino24 sheds light on the ambiguous legal standing surrounding ownership of

18 Pamela Samuelson, supra note 14.

19 Copyright Act, 1957, § 51, No. 14, Acts of Parliament, 1957 (India).

20 Indian Performing Right Society v. Eastern India Motion Pictures Association, (1977) AIR 1443.

21 Digital Millennium Copyright Act, Pub. L. No. 105-304, 112 Stat. 2860 (1998) (U.S.A.).

22 Aaron Wright and Primavera De Filippi, Blockchain and the Law

The Rule of Code 123-125 (Harvard University Press 2019).

23 Hermès International v. Rothschild, No. 1:22-cv-00384 (S.D.N.Y. 2023).

24 Miramax v. Tarantino, No. 2:21-cv-08979 (C.D. Cal. 2022).

digital assets and copyright claims in transactions involving NFTs. De Filippi and Wright further postulate in their book that smart contracts could revolutionize copyright enforcement by embedding automatic licensing and royalty payments in digital assets. Yet enforcement mechanisms on the blockchain carry with them a requirement for an international consensus on regulation, something that is not yet in sight. 25 Accordingly, a hybrid of blockchain technology with traditional copyright registration and enforcement mechanisms may represent a workable way ahead. The Metaverse in the coming years may indeed witness the adoption of more procedural approaches to solve enforcement problems through legal innovation, international cooperation, and technological adaptation.

If this does not happen, the rampant copyright infringement would hinder creative industries and discourage investments in AI-generated content. DATA BIAS AND ETHICAL IMPLICATIONS OF AI TRAINING The deployment of AI in all stages of creativity and decision-making, in this particular case dealing with AI-generated content in the Metaverse, raises substantial ethical issues regarding data bias and fairness. The generation of outputs by AI systems requires extensive datasets, which may include historical biases contained in the source material. The inheritance of such biases places AI-generated works in a dimension where some groups may be disenfranchised, leading to some serious ethical and legal consequences. Perhaps one of the most important issues regarding bias in AI training data is insufficient diversity within datasets.

Studies have shown that datasets collected mainly from Western countries, in English, and with male authorship skew AI outputs. 26 This issue becomes even more serious in the case of the Metaverse since AI-based avatars, avatars in virtual assistants, and automated content generation tools could either reinforce stereotypes or poorly represent minority groups. Bias in AI content may have legal effects, giving rise to discrimination claims and unfair treatment under current anti-discrimination law. In the case of State of Missouri v. Jenkins27, the U. S. Supreme Court noted the inherent dangers that systemic bias could pose on

25 Aaron Wright, supra note 22.

26 David Mittelstadt et al., The Ethics of Algorithms

Mapping the Debate, 3 Big Data & Soc’y 1, 5-8 (2016).

27 State of Missouri v. Jenkins, 515 U.S. 70 (1995).

institutional decision-making. The case highlighted biases in education, but its ramifications extend to governance of AI, in which trained AI systems with biased datasets could deepen similar structural inequities. In a similar vein, the Wisconsin Supreme Court addressed systemic concerns surrounding algorithmic biases in State v. Loomis28, especially concerning the transparency and fair functioning of AI-driven risk assessments at sentencing and thus called for greater accountability and further oversight of AI applications affecting basic rights. The opacity of machine learning algorithms, or 'the black-box problem', is another important issue in AI ethics.

Users and regulators alike are often left in the dark as to the reasoning by which AI systems arrive at their decisions, thereby complicating any inquiries into whether whatever work was produced by AI stands in violation of copyright and ethical principles. The operational dimensions of this problem become magnified in the Metaverse, wherein AIs are capable of executing transactions, generating digital assets, and moderating users. Legal scholars such as Frank Pasquale in “The Black Box Society: The Secret Algorithms That Control Money and Information” believe that, to mitigate against some of these risks, an AI system should always be designed for explainability and transparency.

29 A potential solution to address bias and ethical concerns in AI training is the implementation of "algorithmic auditing"—a process by which AI models are systematically evaluated for fairness and bias before deployment. The European Union’s Artificial Intelligence Act30 proposes mandatory auditing requirements for high-risk AI systems, including those used in intellectual property creation and enforcement. The act sets a precedent for the regulation of AI to ensure ethical and legal standards are followed in the content it generates. Ultimately, regulations, technologies, and industry best practice mechanisms have to be set in place for addressing ethical and legitimate concerns with respect to AI biases arising in the Metaverse.

The policymakers should, in the meantime, make joint efforts to have universally accepted standards for AI fairness and accountability; and the developers should in the selection of datasets promote transparency and inclusivity. If these steps are not taken, then AI content-generation processes may potentially aggravate existing historical injustices and

28 State v. Loomis, 881 N.W.2d 749 (2016).

29 Frank Pasquale, The Black Box Society

The Secret Algorithms That Control Money and Information 1-18 (Harvard University Press 2015).

30 Artificial Intelligence Act, 2024, OJ L, 2024/1689 (EU).

compromise the even-handedness of intellectual property rights distribution in this digital age. With the ascendancy of AI in the Metaverse, it becomes critical that AI-generated works afford consideration to many perspectives and ethical standards. Otherwise, systematic discrimination, legal quandary, and distrust within AI-driven creative processes might arise. This wisdom sets the tone for the necessity of creating a legal and technological framework for AI-generated work, which would focus on fairness, accountability, and inclusivity. PROPOSED SOLUTIONS AND THEIR IMPACT The interplay of AI, the Metaverse, and copyright laws create great complexities that require a multifaceted response embracing legal, technological, and policy considerations. To address the authorship issues, enforcement gaps, and ethical issues, the following solutions are proposed:

1. Redefining Authorship in AI-Generated Works

A major legal challenge with AI-generated content is the lack of any recognizable author. As previously mentioned in the essay, one possible avenue is to create a paradigm of "AI-assisted authorship," where the human presence is significant.31 Copyright ownership can be vested in the party controlling the AI system just as corporations hold copyrights for works prepared by their employees under the work-for-hire principle in the U.S. Copyright Act32. The U.K. Copyright, Designs and Patents Act, 1988 offers a precedent for such a model, as Section 9(3)33 states that the person who makes the necessary arrangements for a computergenerated work is considered the author. Implementing a similar provision in Indian copyright law under Section 2(d) of the Copyright Act, 195734, would provide much-needed clarity and legal certainty.

2. Strengthening Copyright Enforcement in the Metaverse

The key characteristics of the Metaverse decentralization greatly impede its enforcement. 31 Report on Intellectual Property Rights for the Development of Artificial Intelligence Technologies, European Parliament (2020), Motion for a European Parliament Resolution, ¶ J, 15. (supra note 15)

32 U.S. Copyright Act, 17 U.S.C. § 101 (U.S.A.).

33 Copyright, Designs and Patents Act, 1988, c. 48, § 9(3) (U.K.).

34 Copyright Act, 1957, § 2(d), No. 14, Acts of Parliament, 1957 (India). (supra note 8)

Traditional enforcement mechanisms, as discussed earlier, do not find much utility in virtual worlds, where statutory laws and cease-and-desist letters are often ineffective. The real potential lies in using blockchain technology for copyright management.35 Smart contracts executed on blockchain networks can automate copyright licensing and guarantees for fair distribution of royalties to right holders. Such a system will allow for realtime tracking of digital assets and prevent unauthorized reproduction and ensure compliance with copyright laws. Blockchain-based rights management, therefore, corresponds to worldwide initiatives such as the World Intellectual Property Organization discussions on AI and Copyright Enforcement.36

3. Creating an International AI Copyright Treaty

The transnational nature of AI-generated content and transactions occurring in the metaverse aggravates the need for cooperation on an international level. While the Berne Convention and the Trade-Related Aspects of Intellectual Property Rights (TRIPS) form the broad scheme of copyright enforcement, many fundamental issues related to AI have been left untouched. Daniel J. Gervais opines that an international framework is necessary to create uniform copyright standards with regard to AI-created works with the intent to curtail jurisdictional inconsistencies and enhance cross-border enforcement.37

4. Ethical AI Governance and Fair Use Reforms

A comprehensive solution must also consider ethical concerns related to AI bias and fair use.

Implementing algorithmic auditing requirements, as proposed in the European Union’s Artificial Intelligence Act38, would ensure that AI models used in creative processes meet fairness and transparency criteria. In addition, extending fair use exceptions to apply to datasets used for training AI, as discussed in Google LLC v. Oracle Am., Inc.39, could generate some sort of understanding between copyright owners and AI developers to promote innovations

35 Aaron Wright and Primavera De Filippi, Blockchain and the Law

The Rule of Code 123-125 (Harvard University Press 2019). (supra note 22)

36 The WIPO Conversation on Intellectual Property and Artificial Intelligence (2020).

37 Daniel J. Gervais, International Intellectual Property

A Handbook of Contemporary Research 121 (Edward Elgar Publishing Ltd. 2015)

38 Artificial Intelligence Act, 2024, OJ L, 2024/1689 (EU). (supra note 30)

39 Google LLC v. Oracle Am., Inc., 141 S. Ct. 1183 (2021).

while also keeping in mind IP rights protection. These legal, technological, and policy-oriented solutions can lead to a strong copyright framework that encourages innovators while also providing for the rights of the creators. Such reforms will create not just an avenue for improving legal clarity but also one for ensuring that AI output within the Metaverse works under a coherent, moral, and enforceable intellectual property climate.

CONCLUSION

The advent of AI and the rapid expansion of the Metaverse are creating major challenges against traditional doctrines of copyright, which therefore require an immediate overhaul of legislation and regulations. Existing laws such as the Copyright Act of 1957 and international treaty systems like that of the Berne Convention and the TRIPS Agreement appear ill-equipped in dealing with the challenges caused by AI-generated content and decentralized virtual spaces. In the absence of a strong regulatory framework, uncertainty as to the ownership and enforcement of intellectual property rights and ethical considerations may severely hinder innovation and economic growth in these emerging areas. A pivotal conclusion in the present analysis is the dire necessity for the redefinition of authorship concerning AI-generated works. The judgment in Feist Publications, Inc. v.

Rural Telephone Service Co. 40, which asserted that originality is a requirement for copyright protection, may be a threshold that AI-generated pieces could fail to cross under current interpretations of the law. Also, this reluctance in denying the capability of an AI system being regarded as an autonomous creator has been reaffirmed in the Thaler v. Comptroller-General of Patents, Designs and Trademarks41 judgment. This judicial reluctance calls for the urgent requirement of legislative intervention to give definition and structure to the rights and duties of AI developers, users, and other participants. Copyright law enforcement in the Metaverse remains an unsolved conundrum. The case of Indian Performing Right Society v. Eastern India Motion Pictures Association42 elucidates the requirements of flexible forms of enforcement for digital spaces. Blockchain, as suggested by

40 Feist Publications, supra note 5.

41 Thaler, supra note 6.

42 Indian Performing Right Society, supra note 20.

Primavera De Filippi, could be integrated to develop a transparent and decentralized system for copyright management so that digital assets in the Metaverse can be protected against unauthorized copying and infringement. 43 Legal issues are further complicated when considering the morally charged matter of AI training data. The implications of the use of biased datasets in AI-generated content are that discrimination may be propagated, and this may lead to calls for mechanisms of oversight, such as algorithmic audits. The Artificial Intelligence Act44 takes the further step of advancing regulation focusing on transparency and accountability, attaching an even greater possibility of fairness-based analysis being adopted by the courts in the fashion of State v. Loomis45 to have AI-generated works not underscored by systemic biases.

Thus, forced by the multifarious nature of the challenges, there is an urgent need for multiple remedies. Some of these include reforming copyright legislation with respect to dependent AIassisted works, following models such as the U. K. 's Copyright, Designs and Patents Act, 198846, which acknowledges authorship to whoever undertakes appropriate measures in relation to computer-generated works. Second, enforcement mechanisms should be improved in order to avail technological means like blockchain solutions for copyright management and smart contracts to track the use of digital assets in the Metaverse in real-time. Third, the culture of ethical and regulatory oversight needs to be strengthened to minimize bias and promote inclusivity in AI-generated works.

In conclusion, the use of AI in creative industries and the development of the Metaverse are paradigmatic opportunities that require tremendous grasp. Thus, with a bit of forethought on influencing technological innovation and new laws, policy-makers could introduce equilibrium and equity in the intellectual property regime able to safeguard the interests of creators while providing the necessary incentive for promoters to put in effort in their trade. One might think that the future of AI content and digital ownership in the Metaverse will depend on whether legal regimes can keep pace with technological change to ensure sustainability and fairness in the sphere of digital economy sheltered by legal coherence.

43 Aaron Wright and Primavera De Filippi, Blockchain and the Law

The Rule of Code 123-125 (Harvard University Press 2019). (supra note 22)

44 Artificial Intelligence Act, 2024, OJ L, 2024/1689 (EU). (supra note 30)

45 State v. Loomis, 881 N.W.2d 749 (2016). (supra note 28)