Authored By: Parul
Panjab University
ABSTRACT:
Generative Artificial Intelligence is fundamentally reshaping fashion design by independently producing collections, patterns, and textiles that challenge the human-centered foundations of copyright law. This article undertakes a comparative legal analysis across five jurisdictions, namely India, the United Kingdom, the European Union, the United States, and Ireland, illustrating a fragmented global landscape ill-equipped to govern AI-generated creative output. Through critical examination of the DABUS litigation, Zarya of the Dawn, and India’s RAGHAV AI controversy, the article interrogates unresolved tensions surrounding authorship attribution, originality thresholds, and ownership allocation. It further identifies Ireland’s statutory framework as uniquely conflicted within the broader EU copyright doctrine. The article concludes that ad hoc jurisdictional responses are insufficient for an industry where intellectual property constitutes a core element of commercial identity. It therefore proposes a reformed framework centered on authorship transparency, training-data accountability, and adaptive protection mechanisms capable of addressing the challenges of machine-assisted creativity.
KEYWORDS: Artificial Intelligence, Fashion Law, Luxury Law, Copyright, Authorship, Intellectual Property, Fashion Design, AI-Generated Works.
INTRODUCTION:
Fashion functions as both a cultural artifact and an economic product, communicating identity, social values, and creative expression through design. Traditionally, this creative process has been associated with human designers whose artistic choices form the basis of legal protection under copyright law. For centuries, the designer stood at the center of this process, translating imagination into wearable expression. That centrality is now being challenged. According to the World Intellectual Property Organization, generative AI systems can independently analyze vast fashion datasets and produce original designs, textile patterns, and complete collections with negligible human direction, fundamentally altering the question of who, or what, creates fashion.
This shift carries consequences far beyond the studio. Modern copyright systems were developed on the assumption that creative works originate from human intellectual effort. The emergence of generative artificial intelligence challenges this assumption by enabling machines to produce outputs that resemble human artistic expression. When a human designer sketches a garment, the law knows what to do. When an algorithm generates one autonomously, it does not.
The problem is not merely theoretical. Luxury houses, independent designers, and emerging fashion startups are already deploying AI tools in commercial workflows. Yet the legal infrastructure governing ownership, authorship, and infringement has not kept pace. Who owns an AI-generated print? Can it be registered? Can it be enforced against a copyist?Across jurisdictions, these questions produce different and often contradictory answers.
This article examines how five jurisdictions, namely India, the United Kingdom, the European Union, the United States, and Ireland, currently address copyright in AI-generated fashion design. It analyzes landmark cases including the DABUS litigation, Zarya of the Dawn, and India’s RAGHAV AI controversy to identify where existing frameworks succeed, where they fail, and where they remain dangerously silent. Ultimately, it argues that the fashion industry’s growing dependence on generative AI demands not incremental judicial interpretation but deliberate, forward-looking legal reform.
ARTIFICIAL INTELLIGENCE, COPYRIGHT AND OWNERSHIP IN FASHION DESIGN:
The integration of generative AI into fashion has moved beyond experimentation; it now actively shapes design creation, trend forecasting, and consumer engagement at a commercial scale. AI systems autonomously generate garment designs, textile prints, and complete collections by processing enormous volumes of fashion data, enabling brands of all sizes to accelerate creative output significantly.
Yet this capability creates a fundamental legal rupture. Copyright law was architected around human creative agency, treating authorship and originality as inseparably human qualities. As Samuelson observed, allocating ownership rights in computer-generated works presents challenges that traditional copyright doctrine was never designed to resolve. When AI generates a design independently, competing ownership claims emerge simultaneously, from the operator deploying the system, the developer who built it, and the brand commercializing the output.
Most legal systems currently demand meaningful human authorship before extending protection, leaving AI-generated fashion designs in a state of legal vulnerability that designers, brands, and investors cannot afford to ignore.
INTERNATIONAL LEGAL APPROACHES TO AI GENERATED WORKS:
The global legal response to AI-generated works has been neither uniform nor coordinated.Each jurisdiction approaches the authorship question through the lens of its own doctrinal tradition, producing outcomes that are inconsistent, difficult to reconcile, and increasingly inadequate for a commercially integrated fashion industry.
United States
The United States adopts one of the most restrictive approaches toward AI-generated works by requiring demonstrable human authorship as a prerequisite for copyright protection. The U.S. Copyright Office’s decision in Zarya of the Dawn (2023) crystallized this approach when it refused protection for AI-generated images produced through Midjourney, while simultaneously protecting the human-authored text and arrangement surrounding them.The Office reasoned that copyright exists exclusively to incentivize and reward human creative expression. For AI-generated fashion designs, this creates an immediate commercial problem; regardless of a design’s originality, aesthetic sophistication, or market value, it will not attract copyright protection in the United States if meaningful human creative input is absent. Brands operating in the American market therefore cannot rely on copyright as a primary enforcement mechanism for AI-generated outputs.
United Kingdom
The United Kingdom departs from the traditional human-authorship model by providing a statutory mechanism for assigning authorship in certain computer-generated works. Section 9(3) of the Copyright, Designs and Patents Act 1988 provides that authorship of computer-generated works belongs to the person who made the arrangements necessary for their creation. Rather than allowing AI-generated works to fall into the public domain by default,Parliament anticipated, even if imperfectly, the possibility of non-human creative output decades before generative AI became commercially significant. In fashion contexts, this provision potentially assigns ownership to the designer, creative director, or fashion house that configured and directed the AI system. However, the provision’s age creates interpretive uncertainty; courts have yet to determine what level of human arrangement satisfies the statutory threshold when modern generative AI operates with considerable autonomy.
European Union
The European Union occupies a firmly human-centered position, anchored by the Court of Justice’s ruling in Infopaq International v Danske Dagblades Forening (2009), which established that protected works must constitute the author’s “own intellectual creation,”reflecting personal creative choices and human personality. Purely AI-generated designs cannot satisfy this standard. However, the EU framework leaves meaningful room for AI-assisted works; where a human designer exercises genuine creative control over the AI’s output, selects from generated options, and shapes the final result, copyright protection may still be available. For luxury fashion houses operating across EU member states, this distinction between AI-generated and AI-assisted design is not merely academic; it determines whether their creative investments attract legal protection at all.
India
India currently presents the most uncertain landscape among the jurisdictions examined.The Copyright Act, 1957 contains no provision addressing AI-generated works, and Indian courts have not yet confronted the question directly. The Supreme Court’s decision in Eastern Book Company v D.B. Modak (2008) established that originality requires skill,judgment, and a minimum degree of creativity, rejecting mere labor as sufficient. While this standard implies that human intellectual contribution remains essential, it provides no guidance on how to assess works where AI performs the majority of creative labor. India’s rapidly expanding fashion and technology sectors make this silence increasingly costly.Designers and brands investing in AI-generated collections currently operate without any reliable indication of whether their outputs are protectable, enforceable, or vulnerable to imitation.
Ireland
Ireland presents perhaps the most structurally distinctive approach. The Copyright and Related Rights Act 2000 explicitly recognizes computer-generated works and assigns authorship to the person who undertook the arrangements necessary for their creation, mirroring the UK model but enacting it within a distinct legal tradition. In practice, this may extend to individuals or organizations that configured, operated, or directed the AI system producing a fashion design.
However, Ireland’s position is complicated by its EU membership. While Irish statute assigns authorship to a human actor even where the creative work is machine-generated, EU copyright doctrine requires that a work reflect genuine human intellectual creation. This creates a genuine normative conflict; a design protected under Irish law may simultaneously fail to qualify for protection under the broader European framework within which Irish law must operate. Future litigation will likely be required to resolve this tension, leaving Irish fashion businesses in a position of legal ambiguity in the interim.
Comparative Assessment
A comparative review demonstrates substantial divergence in national approaches to AI-generated creativity, resulting in legal uncertainty for businesses operating across multiple markets. The United States and European Union prioritize human creativity and largely exclude purely AI-generated works from protection. The United Kingdom and Ireland provide statutory mechanisms for assigning authorship to responsible human actors, though both frameworks face interpretive challenges in the context of modern generative AI. India remains without legislative or judicial guidance entirely.
For luxury fashion brands operating internationally, this fragmentation is not an abstract concern; it directly affects licensing strategy, enforcement capacity, and investment decisions. An AI-generated design may be protectable in London, unprotectable in New York, and legally ambiguous in Mumbai simultaneously. Until jurisdictions develop a more coherent and coordinated approach, the international fashion industry will continue navigating an inconsistent and unpredictable intellectual property environment.
CASE LAW ANALYSIS: THE EVOLUTION OF AUTHORSHIP AND AI GENERATED CREATIVITY:
Courts have played a significant role in defining the limits of copyright protection in technologically mediated creative processes, consistently emphasizing the importance of human intellectual contribution.
In Feist Publications v Rural Telephone Service (1991), the U.S. Supreme Court rejected the “sweat of the brow” doctrine, holding that creative contribution, rather than labor intensity, forms the qualifying threshold for copyright protection; a standard that AI-generated fashion outputs struggle to satisfy by definition. When algorithms autonomously generate thousands of design variations, none can be straightforwardly credited with the human creativity Feist demands.
The EU reinforced a comparable standard in Infopaq International v Danske Dagblades Forening (2009), requiring that protected works constitute the author’s “own intellectual creation” reflecting personal creative choices. Purely automated fashion outputs, however sophisticated, struggle to satisfy this standard, effectively excluding them from European copyright protection unless meaningful human creative direction is demonstrated.
India adopted a similar position in Eastern Book Company v D.B. Modak (2008), where the Supreme Court held that originality requires skill, judgment, and creativity beyond mere labor. This reasoning was practically tested through the RAGHAV AI controversy, where a copyright application naming an AI system as co-author was challenged precisely because Indian law does not recognise AI as a legal author. No judicial ruling followed, but the episode exposed an urgent legislative gap, one that India’s growing AI-driven fashion sector cannot afford to leave unaddressed
The non-human authorship question was sharpened in Naruto v Slater (2018), where the court held that legal personhood, not mere creative output, determines copyright eligibility. Although involving a monkey rather than a machine, the decision’s reasoning has become foundational in AI authorship debates, firmly closing the door on non-human copyright claims under existing frameworks.
The DABUS litigation, Thaler v Comptroller-General of Patents, Designs and Trade Marks (2021-2023), extended this principle into intellectual property more broadly. The UK Supreme Court’s conclusion that only natural persons can hold inventor status, though decided in patent law, carries significant implications for copyright, reinforcing the view that AI cannot independently possess or assert legal rights over its creative outputs.
Zarya of the Dawn (2023) produced the most directly applicable ruling for fashion contexts. The U.S. Copyright Office’s distinction between protectable human-authored elements and unprotectable AI-generated images established a practical framework; AI-assisted works may attract protection where humans exercise genuine creative control, but autonomous AI outputs generally do not.
In Hermès International v Rothschild (2023), the court confirmed that intellectual property rights remain fully enforceable in digital environments. Ireland’s Copyright and Related Rights Act 2000 similarly anticipates machine-generated authorship by assigning it to the person who made necessary arrangements for the work’s creation. However, Irish courts have yet to interpret this provision against modern generative AI, and its compatibility with the EU’s human intellectual creation standard remains unresolved, leaving Irish fashion businesses in a position of genuine legal uncertainty.
Collectively, these decisions reveal an unambiguous judicial trajectory; copyright law continues to prioritize human authorship, and no jurisdiction has yet extended autonomous protection to AI-generated creative output.
CHALLENGES FOR LUXURY BRANDS AND THE NEED FOR LEGAL REFORM:
The luxury fashion industry derives significant value from perceptions of originality, craftsmanship, and designer authorship. The increasing use of generative AI raises important questions regarding the preservation of these characteristics within existing intellectual property frameworks. As AI assumes a greater role in creative processes, traditional assumptions about authorship, authenticity, and ownership become increasingly difficult to sustain.
The most immediate threat concerns creative identity. Consumers purchasing luxury fashion invest in a narrative, the heritage, craftsmanship, and artistic personality of a designer or house. When AI contributes substantially to the design process without disclosure, that narrative becomes legally and ethically compromised. Brands like Gucci and Balenciaga, which have publicly experimented with AI-generated visuals, already face the question of whether their AI outputs can be registered or enforced under current intellectual property regimes. Undisclosed AI involvement additionally risks consumer protection challenges, particularly where brand identity is marketed as an expression of human artistry.
Training data presents a subtler but equally serious vulnerability. Generative AI systems learn from vast repositories of fashion imagery, runway archives, and design catalogues, much of which carries existing intellectual property protection. Luxury brands occupy a uniquely exposed position, functioning simultaneously as potential users of AI systems and as victims of them. An AI trained on a brand’s visual archive can reproduce its aesthetic language with sufficient variation to evade direct infringement claims, yet close enough to dilute brand distinctiveness and facilitate imitation at industrial scale.
Enforcement complexity compounds these challenges. AI-generated designs that closely approximate iconic luxury products, remaining just sufficiently different to resist clear infringement claims, are increasingly difficult and expensive to pursue. The MetaBirkins dispute demonstrated that enforcement in digital environments is possible but resource-intensive, requiring constant vigilance that smaller designers simply cannot sustain.
These converging pressures demand urgent legislative response. Current copyright frameworks offer luxury brands, independent designers, and consumers no reliable guidance on ownership attribution, training data liability, or disclosure obligations. Reform must address three priorities: establishing clear authorship and ownership rules for AI-generated outputs, introducing accountability mechanisms for the use of copyrighted works in AI training, and mandating transparency where AI significantly contributes to commercial creative products. Without such reform, the fashion industry’s most valuable asset, authentic human creativity, risks being progressively undermined by the very technology it has embraced.
CONCLUSION:
The growing integration of generative AI into fashion design has exposed fundamental limitations within existing copyright frameworks. A comparative analysis of the United States, the United Kingdom, the European Union, India, and Ireland demonstrates that significant uncertainty persists regarding authorship, ownership, and the protection of AI-generated works. While most jurisdictions continue to prioritize human creativity as the basis of copyright protection, none provides a fully satisfactory response to the challenges posed by increasingly autonomous creative technologies.
This uncertainty carries practical consequences for designers, fashion houses, technology developers, and consumers alike. Questions concerning ownership of AI-generated outputs, the lawful use of copyrighted training data, and transparency in AI-assisted design remain inadequately addressed under current legal regimes.
Accordingly, future regulatory efforts should move beyond reactive, case-by-case adjudication and towards comprehensive legislative reform. Such reform should establish clear standards for authorship attribution, introduce accountability mechanisms for training-data use, and promote transparency where AI materially contributes to commercial creative products. As technological innovation continues to reshape the fashion industry, copyright law must evolve in a manner that balances the protection of human creativity with the realities of machine-assisted design.
BIBLIOGRAPHY:
CASE LAW
1.Eastern Book Company v. D.B. Modak, 2008, Supreme Court of India, (2008) 1 SCC 1.
2.Feist Publications Inc. v. Rural Telephone Service Co., 1991, Supreme Court of the United States, 499 U.S. 340.
3.Hermès International v. Rothschild, 2023, United States Court of Appeals for the Second Circuit, 83 F.4th 1248.
4.Infopaq International A/S v. Danske Dagblades Forening, 2009, Court of Justice of the European Union, Case C-5/08.
5.Naruto v. Slater, 2018, United States Court of Appeals for the Ninth Circuit, 888 F.3d 418.
6.Thaler v. Comptroller-General of Patents, Designs and Trade Marks, 2023, United Kingdom Supreme Court, UKSC 49.
7.Zarya of the Dawn, 2023, U.S. Copyright Office, Registration Decision.
STATUTES
1.Copyright Act, 1957, India.
2.Copyright and Related Rights Act, 2000, Ireland.
3.Copyright, Designs and Patents Act, 1988, United Kingdom.
JOURNAL ARTICLES
1.Samuelson, Pamela, “Allocating Ownership Rights in Computer-Generated Works,” University of Pittsburgh Law Review, Vol. 47(4), 1986, pp. 1185–1228.
2.Bonadio, Enrico and McDonagh, Luke, “Artificial Intelligence as Producer and Consumer of Copyright Works: Evaluating the Consequences of Algorithmic Creativity,” Intellectual Property Quarterly, 2020.
3.Scafidi, Susan, “Intellectual Property and Fashion Design,” Research Handbook on Intellectual Property and Fashion Law, 2019.
4.Aplin, Tanya, “Copyright, Artificial Intelligence and Creative Works: A Comparative Perspective,” European Intellectual Property Review, 2021.
WEBSITES
1.World Intellectual Property Organization, “WIPO Technology Trends 2019: Artificial Intelligence,” WIPO, https://www.wipo.int/tech_trends/en/artificial_intelligence, Accessed: 06 June 2026.
2.U.S. Copyright Office, “Copyright and Artificial Intelligence,” U.S. Copyright Office, https://www.copyright.gov/ai, Accessed: 06 June 2026.
3.European Parliament, “EU AI Act: First Regulation on Artificial Intelligence,” European Parliament, https://www.europarl.europa.eu, Accessed: 06 June 2026.
4.Thakur, Ananya, “Legal Implications of AI-Generated Works in Copyright Law: An Analysis of Raghav Artificial Intelligence v. Union of India,” Lawful Legal, https://lawfullegal.in/legal-implications-of-ai-generated-works-in-copyright-law-an-analysis-of-raghav-artificial-intelligence-v-union-of-india/, Accessed: 06 June 2026.
