October 9, 2020
With the technological advancements being developed by artificial intelligence (“AI”), a branch of computer science wherein computers and other machines simulate human intelligence processes, AI will provide an avenue for furthering innovation and emerging technologies based upon AI and AI-driven processes. The prevalence of AI is transforming the methods in how companies create products, interact with suppliers and customers, and predict trends, and is being viewed as an increasingly useful tool in a number of industries. One industry that has taken advantage of the use of AI is the fashion industry. The fashion industry has been able to utilize AI in a number of innovative ways through the compilation of data from the Internet and social media to gain insight on quickly evolving fashion trends, identifying what consumers want, assisting in the design process, and collecting data on key features of competing fashion lines and price points. The access to such large amounts of data and the ability to synthesize that data in an objective and efficient manner is an advantage to companies in the fashion industry, but these developments, even though seemingly beneficial, may create new legal challenges and threats to traditional intellectual property rights and implicate privacy concerns in a way that was not traditionally problematic.
Because the use of AI may raise practical challenges for businesses regarding the ownership of intellectual property rights, the line between human creation and machine-created technologies may become blurred. Understanding the potential legal implications in using artificial intelligence is critical and not something to be overlooked. With these developments, the analysis and commentary on potential implications of the use of AI-generated works is becoming a trending topic among both lawyers and various industries, as this new area of technology creates unique legal questions, namely with respect to intellectual property and privacy issues. Despite these potential legal issues and uncertainties regarding AI, the fashion industry will likely to continue to use AI, as the benefits of AI allow brand owners to remain competitive and will likely play a more expansive role in product development.
Intermingling the Use of Artificial Intelligence and Human Capabilities in the Fashion Industry: Uses of Artificial Intelligence by Fashion Companies
The fashion industry is one example of where AI has been particularly useful in being able to gather and analyze large amounts of data to better address what consumers want. In the age of the Internet and social media, it is more challenging for designers, retailers, and suppliers to predict what consumers want and how to address consumers’s wants in a expeditious manner. Trends in the fashion industry are more rapidly changing with consumer access to the Internet and social media, and not being able to monitor these changes can be a hindrance to companies in this industry where humans cannot possibly process all of the data from individual consumers and entire markets, social media feeds, and customer product reviews. Additionally, it is difficult for humans to not be influenced by their own bias and thought processes.
This is where AI has come in and served to address the fashion industry’s need to predict what individual consumers will buy and quickly monitor trends that the broader public will buy. AI, combined with data available via the Internet and social media, now allows the fashion industry to collect and analyze large amounts of data, which is revolutionary and allows companies to be more responsive to the market and desires of consumers. For example, Burberry, a global and well-known luxury fashion brand, has used big data and AI in order to enrich its customer and sales relationships and as a tool to combat counterfeit products. The company uses big data by encouraging customers through loyalty and reward programs, and Burberry uses this data to offer tailored recommendations for both online and in-person shoppers and as a method to increase sales. Additionally, Burberry is using big data and AI to combat counterfeiters and identify knock-off products through an AI-powered image recognition technology from Entrupy. This technology can determine whether a product is authentic or fake and claims a large percentage of accuracy (99%) in identifying counterfeit items. As a company that is one of the most counterfeited labels in the world, such technology is incredibly valuable to a company like Burberry and allows Burberry to quickly shutdown and implement enforcement measures against counterfeiters.
Another example of companies using AI to deliver innovative experiences to its customers is the creation of “virtual stylist” by Stitch Fix. Relying upon customer input and collaboration with AI and human stylists, the “virtual stylist” uses AI to analyze data on style trends, body measurements, and customer feedback to help the human stylists provide a consolidated list of possible recommendations. Additionally, AI technologies have also been used in the design process. An example of this is Tommy Hilfiger’s partnership in January 2018 with the Fashion Institute of Technology (“FIT”) Infor Design and Tech Lab and IBM on a project called Reimagine Retail. The objective of this project was to show how AI technologies could give retailers an edge of speed in the design process and provide the next generation of retail and design leaders with new skills to use AI in the design process. The FIT students created new designs for the Tommy Hilfiger brand using IBM research AI tools and a library of designs from past collections of Tommy Hilfiger designs. The AI provided fabric patterns, colors, and silhouettes that the students incorporated into their designs. Tommy Hilfiger also uses IBM’s AI technology to analyze customer reviews and sales performance for the items offered in its collections, predict upcoming trends, and to aid in the design of its collections.
These examples demonstrate how AI and big data is revolutionizing the fashion industry, but the use of AI and related technologies is not heavily regulated in the United States and poses significant legal implications for intellectual property and privacy issues.
Implications for Intellectual Property Law and Privacy
Despite the benefits of AI for brand owners, the legal implications and issues arising from the growing use of AI in the fashion industry are uncertain and still developing. The use of AI poses legal challenges in the realm of intellectual property rights, where ownership of intellectual property created by or in part by AI is uncertain and where efforts of intellectual property owners can be easily replicated. AI also raises privacy concerns for consumers, where the targeting of individual consumers permeates many categories of data, including personal data, that consumers may not have expected to be the kind of information collected by fashion companies. Companies and their counsel should take into consideration the legal uncertainties of AI, as the development and use of AI increases to maintain a competitive edge in the marketplace.
AI and Copyright Law
Copyright is particular area that is implicated with respect to the interaction of AI and intellectual property rights, as it raises a potential issue of ownership. A key feature of AI being used by the fashion industry that raises copyright questions is the use of AI to generate fabric patterns, colors, and silhouettes. To the extent that AI-generated designs include contributions from both a human and a machine, the question is whether the design is jointly owned or is solely owned by the human who contributed to the design.
At present, there is no clear protection for fashion designs created using AI under U.S. copyright law and no specific law that addresses the ownership of works created by machines. However, the Compendium of U.S. Copyright Office Practices specifically has a human authorship requirement in section 306, which states that “[t]he U.S. Copyright Office will register an original work of authorship, provided that the work was created by a human being. . . . [T]he Office will refuse to register a claim if it determines that a human being did not create the work.” It is not clear how much AI involvement in a design would be required before making a design ineligible for copyright protection, and no current case law in the United States addresses this issue with respect to AI. Notably, the human authorship requirement was recently discussed in a claim of animal authorship made in what is now referred to as the “monkey selfie” case, Naruto v. Slater, where the Ninth Circuit ultimately held that the monkey lacked standing to sue under the Copyright Act as animals are not authorized to file copyright infringement suits under the Copyright Act.
This area of intellectual property will likely be a key issue for the fashion industry with respect to the use of AI technologies in that it may impact the originality threshold required for copyright protection and implicate the human authorship requirement. If such AI-generated designs are not entitled to protection, these designs may more vulnerable to copying and harm fashion companies and brands in the future and raise complicated issues of what can constitute infringement for designs created in part or in whole by AI.
AI and Trademarks and Counterfeiting
Although AI may present legal challenges and uncertainties, AI has offered positive developments for brand owners in the area of trademarks and counterfeiting. Certain AI technologies can offer image processing capabilities that can be used to identify potential infringement and verify the authenticity of a potential counterfeit good. One example, used by Burberry, is a service offered by Entrupy, which offers a machine learning app to detect counterfeit fashion items based upon a database of authentic luxury items and was launched in 2012. This service claims to have a 99.1% accuracy rate and has taken data samples from the last 80 years through the present in order to identify fake versus authentic items. Another service provider, DataWeave, launched a Counterfeit Products Detection solution in 2018 to enable brands to detect and reduce the presence of counterfeit goods on e-commerce websites using AI-powered image and text analytics. DataWeave also offers a suite of products that include retail intelligence services and brand analytics. These two examples of AI-powered services and technologies demonstrate how the use of AI can benefit brand owners, particularly in identifying and curbing infringement and counterfeiting activities across the Internet and around the world.
As these technologies become more widely used by companies in the fashion industry and further refined, AI will be a valuable tool in being able to address and identify infringement for global brands and reduce losses from counterfeiting, while minimizing costs of investigation efforts and enforcement of potential infringers and counterfeiters.
AI and Privacy
An area of concern that has arisen with respect to the use of AI in the fashion industry, and across businesses using AI generally, is the impact AI technologies may have on privacy. Many AI technologies collect a large amount of consumer data that is used by companies to adapt to the market, identify trends, track competitors’ activities via the Internet and social media, and improve consumer experiences and target consumer demands. However, consumers may not be aware of just how much information is being collected from the Internet, and fashion retailers and designers may not be unaware of the privacy law issues that are implicated in the collection, processing, and use of big data and AI.
Although there is no federal regulation or law in the United States that addresses consumer data privacy, a number of states, such as California, have passed consumer privacy laws that impose specific duties on companies or individuals collecting data and also includes provisions that address individual consumer rights. A regulation to be mindful of, particularly for global brands, is the recently enacted General Data Projection Regulation (GDPR) that governs the collection, use, transmission, and security of data collected from any residents of the European Union, which entered into force in May 2018. It is important to be mindful of these kinds of laws and regulations in order to for companies to make certain to put appropriate measures in place and protect consumer data and ensure that consumers give their fully informed consent on all information gathered.
There use of AI and big data has been extremely beneficial in allowing fashion companies to better address market demands, improve relationships with consumers, and use social media and other platforms to increase sales and provide a more interactive experience. Given the rapid development of AI and use of big data, the fashion industry will continue to find new applications for the use of AI. However, the implications and challenges of AI, which is still largely unregulated in the United States, may continue to raise new legal issues, but the law will have to address the intersection of AI with intellectual property and privacy issues to avoid threats to innovation and better define the parameters of ownership and protection of intellectual property and data.