Editor-In-Chief
One AI Outfit, Two Creators: Are We at Copyright’s Breaking Point?
By Aminesh Nayan Gogoi
Editor-in-Chief
Fashion Law IP Blog
As AI-Generated and digital fashion takes off, it raises fresh legal riddles: if an algorithm or a prompt designs a dress, who is the “author”? Must a human add a creative spark for copyright to apply? In short, the old tests of originality and authorship, the minimal “modicum of creativity” versus the old “sweat of the brow” effort-based test, are under stress. This article surveys how different jurisdictions are dealing with AI-assisted fashion, highlights gaps in existing doctrines, and asks if new rules are needed.
Testing the Tests of Originality
Most copyright systems once debated whether mere skill-and-effort (“sweat of the brow”) sufficed for protection or whether a work needed a spark of creativity (“modicum of creativity”). Today, the higher creativity standard prevails. For instance, India’s Supreme Court in Eastern Book Co. v. D.B. Modak explicitly rejected a purely labor-based test in favor of requiring some creative input. The U.S. likewise requires only a “minimal degree of creativity”, discarding any pure labor test. The EU, likewise defines protected works as the author’s “own intellectual creation”, requiring free creative choices and a personal touch. In practice, this means passive copying won’t qualify. In short, most laws today insist on a human creative contribution, not just hard work or data crunching.
Yet AI challenges these assumptions. If a machine autonomously generates a design, is any human creativity involved? Most legal systems still tie authorship to people. In the U.S., the Copyright Office bluntly states that only works “created by a human being” can be registered. In the EU, a work must bear the imprint of a human creator’s personality, an AI acting alone cannot impart that personal touch. And historically, “sweat of the brow” ideas (which allowed protection for diligent compilation even without creative choices) have been abandoned in favor of creativity-based tests. Thus, generative AI’s role, abundant computation without a human “author”, sits uneasily under existing doctrine.
The World, Tailored by AI
Courts and copyright offices across jurisdictions are now being asked a deceptively simple question: can creativity exist without a human behind it?
In the United States, the answer remains firmly grounded in human agency. Copyright law insists on independent human creation and only a minimal spark of creativity, leaving no room for works produced entirely by autonomous systems. The decision in Thaler v Perlmutter makes this position unmistakably clear, where no human author can be found, copyright does not follow. At the same time, regulators stop short of rejecting AI altogether, acknowledging it as a sophisticated instrument rather than a rival creator. When a human meaningfully guides, selects, edits, or curates the output, the law is willing to listen. Creativity, in this view, lies not in the machine’s generation but in the human’s choices.
European copyright law speaks a similar language, but in more introspective terms. A work must reflect the author’s “own intellectual creation,” a phrase that quietly insists on personality, intention, and freedom of choice. A mere button press is not authorship; it is delegation. Where AI operates without genuine human direction, the law sees no personal imprint and therefore no protected work. Yet even here, the ground is shifting. As the European Union debates the implications of generative AI and moves toward disclosure obligations under emerging AI regulation, it tacitly acknowledges that existing categories may no longer be sufficient. The United Kingdom offers an uneasy compromise: when no human author can be identified, copyright is assigned to the person who made the “arrangements” for the work’s creation. This provision functions as a legal stand-in for authorship, though what such “arrangements” mean in an era of prompt engineering and autonomous systems remains conspicuously undefined.
Indian copyright law, too, remains anchored to the human creator. The statute assumes authorship as a human act, defining the author of a computer-generated work as the person who causes it to be created. The Suryast controversy captures this tension vividly. When an artist sought to credit both himself and the AI tool that transformed his photograph, the initial willingness of the Copyright Office to acknowledge this collaboration was swiftly withdrawn. The retreat underscored a deeper discomfort. While AI may participate in creation, it cannot yet be recognised as a legal author. Indian courts have long favoured creativity over mere labour, but AI now complicates even that settled position, prompting renewed calls for legislative reform or even a distinct legal category for AI-generated works.
Elsewhere, the picture is no less fragmented. Chinese courts appear more receptive to extending protection to AI-generated outputs, while jurisdictions like Japan and Singapore remain in exploratory mode, watching rather than deciding. What unites these approaches is not consensus, but hesitation. Across legal systems, authorship continues to orbit around human creative contribution, even as machines increasingly occupy the space once reserved for imagination. Fully autonomous AI designs, for now, remain suspended in a legal grey zone, created, circulated, and consumed, yet still searching for a place within the law’s understanding of creativity itself.
Fabric Ends and Code Begins
These issues are acute in digital fashion. By “digital fashion” we mean computer-created garment designs, 3D models and graphics that can be worn in games, augmented reality or as NFTs. Unlike traditional couture (with many designers, pattern-makers and factories), digital fashion design is often more streamlined: in fact, studies note that digital fashion’s authorship is “more democratic” and closer to the legal idea of a single author. In other words, the person who fashions a virtual dress file is likely the same person seeking copyright, which could make meeting originality easier. The very act of 3D-designing and rendering a garment in software aligns well with the EU standard of an author’s “free and creative choices”.
Yet the AI dimension is largely unexplored. Only a handful of legal analyses even mention digital fashion, and those mostly predate today’s generative AI wave. For example, a recent study of digital fashion designers explicitly sidestepped questions of AI code. It noted that in AI-assisted design, “the designer might write the code and create the method for co-designing with computer and data,” but then admitted that assessing code authorship was beyond its scope. In short, traditional scholarship has not yet dealt with AI tools in fashion. Key questions remain unanswered. When an AI suggests a novel pattern, is the designer who clicked “generate” an author? If a prompt combines thousands of images of runway outfits, is that result an “original” work or an assortment of others’ creativity? Does copying a machine’s output infringe anyone’s rights?
Some specific puzzles illustrate the gap. Fashion law generally teaches that the utilitarian shape or cut of a garment isn’t protected, only separable creative surface designs are. But digital garments have no “utility” in the usual sense. In fact, one analysis observed that digital fashion might escape the strict separability test applied to clothes. Instead, it could be treated like video game art (where the code, graphics and design all get broad protection). If that view prevails, digital/AI designs could enjoy strong coverage. However, no case has directly addressed what constitutes a machine-generated virtual outfit, for example, a 2D motif or a fully coded creation. These uncertainties show how the intersection of AI and digital fashion is a real research gap, scholars have begun mapping digital fashion vs. IP, and have studied AI art in general, but rarely these threads together.
Is it Time for a New Doctrinal Reform?
Given the confusion, critics are asking if old doctrines need patching. Some suggest dual standards: one rule for traditional human-made art, another for AI-assisted work. For example, Friedmann (2025) argues that jurists are already applying a “de facto double standard”, demanding almost “platonic” perfect control from AI users, and proposes formalizing this by requiring AI disclosures and perhaps new rights for AI outputs. Others, especially in India, call for a distinct statutory category or sui generis protection for AI creations, so that neither human authors nor AI companies fall through the cracks. The EU has floated similar ideas, including possibly extending neighbouring or investment rights (currently covering broadcasters or database makers) to AI content producers.
At a minimum, businesses are advised to rely on contracts. Without clear laws, designers using generative tools must document their creative input (kept changes, edits, prompt logs) to show they added a human touch. Platforms’ terms of service can also govern who owns output, some AI vendors grant full rights to users, others do not. But these are temporary solution.
In the longer view, legal systems may need bigger fixes. Across the board, AI on the catwalk is testing the stitch strength of copyright. Does the “modicum of creativity” test really accommodate a designer pressing a button? Is the “sweat of the brow” forever irrelevant, or should we ever credit the massive labor (in data and computation) behind generative models? The debate is just beginning.
The harder question is, are we protecting creativity, or are we protecting our comfort with the idea that only humans can create?
References:
- Eastern Book Company v D B Modak (2008) 1 SCC 1.
- Feist Publications Inc v Rural Telephone Service Co Inc 499 US 340 (1991).
- Thaler v Perlmutter No 1:22-cv-01564 (DDC 2023).
- Ankit Sahni v Registrar of Copyrights (Suryast case) Copyright Office of India (2021–22).
- Copyright Act 1957 (India).
- Copyright, Designs and Patents Act 1988 (UK).
- Copyright Act of 1976, 17 USC (US).
- Directive 2001/29/EC of the European Parliament and of the Council of 22 May 2001 on the harmonisation of certain aspects of copyright and related rights in the information society.
- US Copyright Office, Copyright Registration Guidance: Works Containing Material Generated by Artificial Intelligence (March 2023).
- World Intellectual Property Organization, Revised Issues Paper on Intellectual Property Policy and Artificial Intelligence (WIPO 2020).
- European Commission, Copyright and Artificial Intelligence: Policy Considerations (EU Publications).
- Lionel Bently and others, Intellectual Property Law (5th edn, OUP 2018).
- Jane C Ginsburg, ‘People Not Machines: Authorship and What It Means in the Berne Convention’ (2018) 49(2) Columbia Journal of Law & the Arts 253.
- Eleonora Rosati, Copyright and the Court of Justice of the European Union (OUP 2019).
- Mark A Lemley and Bryan Casey, ‘Fair Learning’ (2021) 99 Texas Law Review 743.
- Susan Scafidi, Who Owns Culture? Appropriation and Authenticity in American Law (Rutgers University Press 2005).
- NYU Journal of Intellectual Property & Entertainment Law, Volume 14, Number 1 (Fall 2024)
- World Intellectual Property Organization, WIPO Lex Database https://wipolex.wipo.int
When the Designer is a Machine, can Indian IP Law keep up?
By Aminesh Nayan Gogoi
Editor-in-Chief
Fashion Law IP Blog
At the moment, fashion is evolving beyond fabric and thread, into data and code. Creativity is no longer just the domain of people who are human, today AI is walking down the runway too. Computer programs that have been trained on everything from conventional fashion history to contemporary consumer behaviour can now spit out unique style pieces of all kinds, whether it be a material print or an entirely virtual outfit. AI driven design tools have made it possible for designers to move beyond old constraints and try new ideas.
The latest development, however, particularly demonstrates a key prompt: Will India’s current intellectual property laws, particularly the Designs Act of 2000, offer protection to fashion products designed by artificial intelligence?
When the Designer is a Machine
Fundamentally, fashion design has always been human labour, influenced by the hands, wisdom, and imaginative decisions of designers. This viewpoint is represented in India’s Designs Act, 2000, which acknowledges the significance of uniqueness in relation to design registration. According to Section 2(d) of the Act, a “design” is any shape, configuration, pattern, or composition applied to an article by any industrial process, as long as it is unique and hasn’t been published before. Although the Act defines a design, it does not specify what constitutes a truly “original” work or whether the person who created it must be a human.
Indian courts, including in Bharat Glass Tube Ltd. v. Gopal Glass Works Ltd. (2008), have consistently held that originality involves some degree of intellectual effort by a human. In the case of AI-generated works, especially those created with minimal human input, this standard creates an interpretative vacuum. If a machine operates autonomously, can it truly be said to exercise intellectual skill or judgment? And if not, who can claim ownership over such outputs?
Who Really Owns AI-Made Fashion?
Who owns an AI-made design? That’s the real challenge. Under current Indian IP law, only natural persons or legal entities count as authors. Due to their lack of legal personhood, AI systems cannot be acknowledged as creators or owners of intellectual property. Real-world issues are brought up by this recognition gap. A fashion design made entirely by an AI with little to no human input, for instance, might not be able to be rightfully attributed to any one person. Such works risk slipping into the public domain, leaving them without copyright protection. In other words, those who invested time and resources in their creation have no protection against misuse or infringement.
Thus, companies using AI-driven tools, fashion designers, and tech developers must navigate a legal maze, especially in cases where it comes to protecting or monetizing these outputs. Other nations, on the other hand, have started to react to this change. According to Section 9(3) of the Copyright, Designs and Patents Act of 1988, the UK grants authorship of computer-generated works to the person who organized their production. Although the question of authorship is still up for debate, China has also made progress by acknowledging the uniqueness of AI-generated works, provided that they exhibit creative merit. However, India still adopts a human-centered strategy that seems to be falling further behind the quickly changing nature of creative production in the fashion sector.
How the World is Dressing AI in Law
Towards gaining perspective, look at how the rest of the world is handling this. The U.S. Copyright Office has made its stance clear, only human-made works qualify for protection in the case of Thaler v. Perlmutter. In the meantime, countries like the European Union and Japan are presently discussing whether they require new or more adaptable legal frameworks to deal with AI-shaped outputs. In order to address the issues of machine-influenced creativity, the World Intellectual Property Organization (WIPO) has launched open discussions through its AI and IP program. This indicates that India must take immediate action. If there aren’t enough legal protections in place, fashion creators using AI may start to hold back. Without clear safeguards, many might hesitate to fully explore what this technology can offer. And if that uncertainty continues, it could end up slowing down innovation, right when the industry is ready to take a big leap forward.
When the Law Lags Behind the Runway
When AI steps into the process, India’s IP law faces its toughest test yet. Clear regulations or organized procedures to handle AI-assisted design applications are conspicuously lacking. For instance, unlike the UK’s Copyright, Designs and Patents Act, which expressly attributes authorship of computer-generated works, India’s Designs Act, 2000 lacks any comparable provision. Moreover, While India has historically lacked settled guidance on AI-authorship, in 2024–2025 the Government and IP offices have convened committees and released draft guidance (IP India’s draft CRI guidance and DPIIT consultations). No final, India-wide statutory change has yet been enacted, so uncertainty remains for AI-created designs. Today, AI-generated work sits in an uncharted and vulnerable space. Why? Because originality and novelty are still judged by outdated standards, rules that assume a human creator.
Apart from causing confusion, this legal uncertainty offers an alarming effect on innovation. From a practical standpoint, ignoring AI in law isn’t just theoretical, it affects independent designers trying to survive. AI tools have opened doors for students, small fashion brands, and designers to compete on a level that once seemed impossible. Now, with minimal funding, one can make some great idea boards, concept-type prototypes, and dreamy images. But if these improvements aren’t legally shielded, then larger corporations can mimic them at no cost, something that ultimately reduces the bargaining power of those who can’t afford to defend themselves.
And that is why our legal system should start to recognize AI-assisted designs. The spirit of creativity and ensuring the fashion industry remains open, dynamic and fair for everyone should be upheld, not simply protecting output.
Virtual Fits, Real Questions
At the nexus of AI and digital fashion, a particularly urgent issue arises. AI-generated clothing is now being tokenized and sold as NFTs. Platforms like Decentraland and Zepeto, they’re turning into virtual fashion hubs, filled with designs born from AI systems.
Luxury brands like Gucci have entered this space with virtual sneakers co-created using AI-informed data analytics. These developments raise several unresolved questions: Do NFTs confer IP rights over the underlying design? Who owns a digital asset created through AI prompts? How does jurisdiction apply when platforms and creators span multiple countries?
In India, the Designs Act does not address digital-only garments or tokenized IP. As the digital fashion market grows, a lack of clarity around ownership and enforceability will hamper its potential.
Our own experience at the Fashion Law IP Blog
At the Fashion Law IP Blog, we have been actively participating in, rather than merely observing, the rapidly evolving intersection of law, technology, and fashion. A dedicated team behind our platform is committed to converting complicated legal developments into understandable, useful information for researchers, designers, and business professionals. We address important topics like traditional fashion and GI, trademark objections, sustainable fashion trends, fast fashion issues, digital fashion and more through well-researched articles, editorials, case studies, comprehensive guides and customizable templates and our consistent “FL Basics” series, breaking down fundamental aspects of fashion and industry related laws. We provide content that is educational but simple to comprehend.
Our workflow behind the scenes incorporates technology and creativity in an effective manner. Even though we use tools like VEED for automatic subtitling, ChatGPT to help draft preliminary scripts based on in-depth human research, and Canva for visual consistency, but every output is shaped, reviewed, and curated by real people. This hybrid approach, where legal reasoning meets automated efficiency, mirrors the very challenges we explore on our journal; authorship, originality, and the blurred boundaries between human and machine.
Can We Stitch AI Into Our Laws?
India needs to move fast if it wants its legal system to keep pace with the growing intersection of law and technology. A good starting point would be revisiting the Designs Act, 2000 to reflect how creativity is changing. This might entail extending the meaning of “authorship” to specifically acknowledge works produced with AI assistance. In addition, India may consider models similar to those in the UK, where the law grants rights to individuals who oversee or manage the AI’s operation; in other words, acknowledging those who work in the background.
Meanwhile, a novel strategy might be taken, introducing a sui generis system designed especially for AI-generated designs. Similar to the EU’s Database Directive, such a system would offer a kind of protection that honours the labour that goes into these innovations without relying on traditional ideas of human authorship.
For real change to happen, we need more than just discussion, we need practical, grounded steps. Agencies like the Indian Patent Office and DPIIT could lead the way by issuing clear, easy-to-follow guidelines. These should explain how to handle AI-influenced design filings, not just for legal experts, but for designers, startups, and everyday creators trying to protect their work in a space that’s still a bit of a legal grey zone.
We also need to talk honestly about credit and contribution. Today, a fashion design might be shaped by a designer’s eye, a coder’s algorithm, a data curator’s inputs, or even the end user’s feedback in a digital environment. To put it in perspective, our laws should start reflecting this layered reality. That doesn’t mean giving a machine a seat at the table, it means making sure that the people behind the machine aren’t erased in the process.
Can the Law Keep Up with the Next Generation of Designers?
The fashion industry is altering rapidly. With the advancement of AI, designers are pushing the envelope, fusing styles, and accelerating the creative process like never before. However, India’s legal system is still working to keep up with the quick changes in technology. Only works that originate directly from the human mind are currently recognized by copyright laws in this country. AI-generated designs are thus left in a grey area with no obvious safeguards.
Some nations are already reconsidering their authorship policies in the era of artificial intelligence. Governments and courts are divided on AI authorship. The UK attributes authorship of “computer-generated” works to the person arranging their creation (CDPA section 9(3)). In the U.S., the Copyright Office and courts stress human authorship, reaffirmed in Thaler v. Perlmutter. China shows mixed views, Feilin v. Baidu did not endorse AI authorship, while Li v. Liu upheld protection where human prompting and selection showed creativity. The global stance remains unsettled and evolving.
If India doesn’t adapt, it risks falling behind. Fast. But embracing AI in law isn’t about abandoning human creativity, it’s about evolving with it. It’s about understanding that creation isn’t what it used to be, and the law can’t afford to stay the same either. As fashion-tech continues to grow in India, protecting new forms of creativity will matter more than ever. Giving proper recognition to AI-assisted designs isn’t just about updating the law, it’s about making sure both human and AI-supported creators have the space to grow, compete, and shine on a global stage.
So, the question is, can we keep up with the future without rethinking who (or what) counts as a creator?
References:
- Designs Act, No. 16 of 2000, Section 2(d), India Code (2000).
- Bharat Glass Tube Ltd. v. Gopal Glass Works Ltd., (2008) 10 SCC 657.
- Copyright, Designs and Patents Act 1988, Section 9(3) (UK).
- Copyright, Designs and Patents Act, 1988.
- Thaler v. Perlmutter, Copyright Office Review Board, U.S. Copyright Office.
- Draft Guidelines for Examination of Computer-Related Inventions (CRI), 2025, Controller General of Patents, Designs & Trade Marks (India), March 25, 2025,https://www.ipindia.gov.in/writereaddata/Portal/Images/pdf/Draft_CRI_Guidelines_Publication_March2025.pdf.
- World Intellectual Property Organization, Revised Issues Paper on Intellectual Property Policy and Artificial Intelligence.
- Fashion Law IP Blog, https://fashionlawipblog.com
- The Fabricant, https://www.thefabricant.com
- Kati Chitrakorn, How Designers Use Generative AI, Vogue Business.
- Ministry of Economy, Trade and Industry (Japan), Intellectual Property Strategy Vision for the Age of AI (2020).
- Beijing Internet Court, Beijing Internet Court Ruling in First Case of Copyright Infringement of AI-Generated Article.
- Li v. Liu, (2023) Jing 0491 Min Chu No. 11279 (Beijing Internet Ct. Nov. 27, 2023)
- Decentraland. (2017). Decentraland (Virtual world). Decentraland Foundation. https://decentraland.org
- Naver Z Corporation. (2018). Zepeto (Mobile application). Naver Z Corporation. https://zepeto.me
The Illusion of Luxury: Is the Dream Worth the Price?
By Aminesh Nayan Gogoi
Editor-in-Chief
Fashion Law IP Blog
The Appeal of Luxury Fashion
Luxury fashion is built on perception and desire. Milan, often referred to as the “fashion capital of the world,” is home to premium labels that span practically every letter of the alphabet, from Armani to Zara. These brands rely on the concepts of exclusivity and craftsmanship, inducing a sense of luxury that draws in customers. Social media also contributes significantly to maintaining this impression. When Jennie Kim’s Alaïa look for the 2024 Met Gala was revealed to have taken 200 hours to craft, or when Jodie Turner-Smith’s Burberry dress crowed 150,000 faux pearls, the internet erupted with admiration. Legacy brands like Prada (founded in 1913), Chanel (1910), and Cartier (1847) have long mastered this game, carefully curating an image of luxury that keeps their status intact.
Luxury in the Indian Market
In India, the relationship between luxury fashion and Bollywood runs deep. Celebrities like Diljit Dosanjh, regarded as India’s most fashionable man, are often seen wearing brands like Balenciaga, Supreme, and Gucci on stage. The popularity of luxury labels is further increased by the frequent mention of their expensive fashion choices in interviews with popular actresses like Alia Bhatt, Priyanka Chopra, and Deepika Padukone. However, luxury fashion is no longer limited to the elite. Influenced by aspirational marketing and social media, the middle class is keen to get involved. Vlogs featuring influencers unboxing designer handbags worth lakhs have increased a growing demand. Even many creators’ content focusses on the subject of which luxury brand of bag is worthwhile purchasing.
Luxury labels like Dior, Louis Vuitton, Gucci, and Prada are particularly popular in India. According to a survey by HypeAuditor, Dior has received the most unique Instagram posts from Indian influencers, resulting in a $26.08 million earned media value in the whole country.
Interestingly, luxury brands today aren’t targeting the older generation as much as they are going after younger consumers. Many millennials and Gen Z buyers willingly spend lakhs on high-end watches, bags, and accessories, often financing them through EMIs. But here’s the question: is the “luxury” marketed to middle-class consumers truly luxury at all?
The Hidden Reality of Luxury Goods
The reality is, luxury brands don’t sell the same experience to everyone. a brand like Gucci might create two rather different dresses: one with subtle, flawless craftsmanship that is worn by celebrities and elite businesswomen, and another with big, eye-catching logos that is aimed for the middle class. Craftmanship, hand tailoring, customized designs and exclusivity, are the domains of true luxury. Yet, the mass-market versions of these designer goods, often priced immensely, rarely justify their cost. Reports have shown that a Dior handbag sold for $2,800 costs just $57 to manufacture. Similarly, Armani pays contractors around $99 per bag, only to sell them for over $1,900 in stores.
That said, behind the glamorous storefronts and sky-high prices, there’s a hidden reality, one that most consumers never see. The very artisans crafting these luxury pieces often work in harsh conditions for minimal pay. While brands proudly highlight “Made in Italy”, or tags, many of these items are actually produced by undocumented migrants, where artisans are paid a fraction of what their work is truly worth, often under exploitative conditions.
This leads to an uncomfortable truth, if luxury is meant to stand for quality and exclusivity, why do so many high-end products resemble mass-produced fast fashion? The answer lies in marketing. Luxury brands sell more than simply clothing or accessories; they sell a dream, a carefully constructed illusion that leads consumers to feel that owning a high-end item is a sign of success and status. Many of these products are really made using cost-cutting techniques while still giving the appearance of luxury.
The Final Question: Who Truly Owns Luxury?
So, while middle-class buyers stretch their budgets, take out loans, or put their savings into a “luxury” purchase, believing they are stepping into an exclusive world, the truth is far different. The real luxury? It remains reserved for a select few.
The question remains, for whom is luxury truly intended?
Is it for the ones who can buy without thinking, or for those who think before buying? Is it the craftsmanship, the exclusivity, or just the price tag that defines it? And if the dream is sold piece by piece, packaged in marketing polish, and made to feel within reach, was it ever truly luxury to begin with?
References:
- HypeAuditor. (n.d.). Dior Emerges as the Top Luxury Brand Among Indian Influencers with $26.08 Million in Earned Media Value. https://hypeauditor.com
- Business of Fashion. (2020, February 24). The True Cost of Luxury: Inside the Sweatshops of “Made in Italy”. https://www.businessoffashion.com
- The New York Times. (2018, December 23). How Luxury Brands Set Prices. https://www.nytimes.com
- Vox. (2019, August 15). Why Designer Bags Cost So Much. https://www.vox.com
- Met Gala 2024 Looks. (2024). Jennie Kim in Alaïa and Jodie Turner-Smith in Burberry. https://www.vogue.com
