The Dance Between Human and AI Fashion Designers

Stella Sun
7 min readSep 7, 2022

A study of how AI-designed clothing is gaining a large foothold in the consumer market, and where will fashion designers fit in this new reality?

Following the golden age of science fiction in the 1940s came Alan Turing’s infamous thought experiment on whether machines can think. Subsequently, in 1955, the first artificial intelligence program was born(1) As our world becomes increasingly digitized, we are relying more heavily on technology for not only production, but also creation.

Throughout the history of large fashion industries, creativity and innovation were often attributed to the great minds of fashion designers who set trends seemingly from their intuition and talent. As technology pervades the fashion world, the production, distribution, and marketing of fashion are gradually overtaken by data. Currently, technology is largely seen as an advanced tool that fashion designers and fashion marketers use to bolster their brands. However, as artificial intelligence permeates from production to ideation, two challenges arise: Will AI-designed clothing gain a large foothold in the consumer market, and where will fashion designers fit in this new reality?

When looking at the trajectory of technology and fashion throughout their shared histories, I anticipate that AI-designed fashion will become a predominant and popular form of fashion consumption in the near future. In the 1970s, computer-aided design, or CAD, was introduced to the fashion industry via footwear design (2). While many CAD programs today are not yet fitted with the power of AI, this technological advancement has allowed fashion designers to generate otherwise unthinkable forms in a virtual three-dimensional world. More importantly, this introduced the idea that machines can work alongside humans in the creation of garments.

An example of a human-machine partnership is Nike’s Flyknit upper for running shoes, which debuted in 2008. This collaboration between human ingenuity and technological expertise reimagined both form and functionality of athletic footwear through materialities achievable only through computerized processes (3).

Nike’s Flyknit shoe upper.

Another predominant push for digitized fashion creation is Iris Van Herpen’s “Capriole” collection in Fall 2011 (4). In collaboration with architect Isaïe Bloch and 3D-printing company Materialise, Herpen’s collection heavily utilized 3D printing and industrialized robotics to achieve her distinguished ethereal motifs (5).

Iris Van Herpen’s “Skeleton” garment from the “Capriole” collection is created from 3D printed nylon.

Furthermore, AI itself has been recently employed in marketing and research for fashion brands. AI marketing is defined as the utilization of “artificial intelligence technologies to make automated decisions based on data collection, data analysis, and additional observations of audience or economic trends that may impact marketing efforts”(6). In addition to internal AI marketing, the rise of using 3rd party AI-run platforms for marketing has also become widespread in the fashion industry. Many labels, such as Zara, H&M, and numerous small e-commerce fashion brands are both selling and reaching larger audiences on social media platforms like Instagram and Tiktok (7). Especially in the present day, these AI-powered social media platforms have proved to be indispensable tools for fashion marketing and retail.

An influencer poses for Instagram and tags the brands that she’s sporting (8).

From material research, mass production, and marketing, technology is now snuggly fit into the success of any fashion brand. Interestingly, AI has also tapped into mainstream fashion creation and is currently working alongside designers to conceive styles, trends, and products.

“How to Generate (Almost) Anything” is a collective of scientists and artists inspired by a course at MIT titled “How to make (almost) anything” (9). In their ventures in AI-produced fashion, they fed around four thousand images of vintage patterns to an AI called the Generative Adversarial Network, or GAN (10). Effectively, the team and its AI software worked together to cultivate an AI fashion designer of vintage garment patterns.

Sampled patterns and AI-designed patterns from “How to Generate (Almost) Anything.”

Another similar AI-human collaboration is the Acne Studios x Robbie Barrat collection. Barrat is an “AI artist,” who uses artificial intelligence to create art and design. In Acne Studios’ Men’s Fall/Winter 2020 collection, thousands of Acne’s archived collections were fed to Barrat’s AI software, and “in the case of Acne’s designs, Barrat’s neural networks were programmed *not* to learn by the correct rules” (11). The results were an entire collection of garments mostly dictated by machines.

Models wearing Acne Studios x Barrat’s Men’s Fall/Winter 2020 Collection.

A perhaps more daring approach to AI-designed clothing comes from the fashion brand Glitch, whose ready-to-wear garments are entirely created by artificial intelligence. Pinar Yanardag and Emily Salvador, the founders of Glitch, were also inspired by the MIT class “How to make (almost) anything” (12). Additionally, their garments are for sale to the consumer market and aim to celebrate women in the STEM field.

Glitch’s AI designed “Little Black Dress.”

In addition to designers benefitting from working with AI, there are also many benefits for the consumer. Tech giants such as Amazon have begun exploration in marketing AI-picked fashion for consumers. Amazon Style Snap is an AI-powered fashion styler that helps you search and style clothes you give to its image search function (13). While this is still in the process of gaining traction, it may play a large role in Amazon’s plans to release its own AI fashion designer (14).

Amazon Style Snap embedded in the Amazon shopping app.

A more established AI shopping platform is Stitch Fix, a website where customers can take a quiz about their styling choices and is sent clothes based on their responses in consultation with millions of other shoppers who completed the questionnaire (15). This widely successful company, which was founded in 2011, carries and retails hundreds of fashion brands in its personal stylist arsenal.

A screenshot of Stitch Fix’s website interface.

Data and technology are already tightly woven into today’s fashion landscape, both for the creators and the consumers. Moreover, the incorporation of artificial intelligence is already being charted as tech companies venture into AI-designed fashion, well-established fashion labels are collaborating with machine learning, and new companies are forming around AI-led fashion. Thus I think that in the near future, AI-designed fashion will gain an increasingly larger grip on the consumer market, which gives rise to the question of the fashion designer’s relationship with the fashion industry.

Presently, there is a mutually beneficial relationship between the AI and the designer. The AI is used lightly as an advanced tool, or exhaustively as a personified member of the team. However, the trajectory of the fashion industry’s relationship with artificial intelligence is aimed at more dependence on the software. Moreover, growing numbers of individuals and companies who are not rooted in the study of fashion design have entered the fashion realm, and are competing with those organically grown from the field. Thus there may be drastic changes in the fashion landscape nearer in the future than we might think or hope.

However, while the roles of fashion designers will most definitely shift, I do not think that they will diminish. Technology is rigid and systemic, while people are emotive and sporadic. I think that fashion designers must grow to work closely with AI, but will be key players in humanizing the software. This may give a new dimension of esteem and elegance for the future fashion designers’ human counterparts.

In the future of AI-designed fashion may also emerge new definitions and connotations for “human-designed fashion.” Just as artisanal craftsmanship became more valued and exclusive as the production line overtook handmade goods, clothes designed by people may gain new appeal as consumers begin to desire that which is more rare. Salvador, the co-founder of Glitch, states, “‘What I think is really cool about these reactive-focus AI tools is that there’s still this really compelling need for a human to intervene with the algorithm’” (12). Perhaps the innate fascination that people have for other people is an important telltale of human ingenuity, creation, and hope that people will find organic connections with each in whichever futures that manifest into reality.


  1. Rockwell Anyoha, “The History of Artificial Intelligence.” Accessed December 11, 2020.
  2. Andy, “How Is CAD Impacting on the Fashion Industry?” Accessed December 10, 2020.,as%20early%20as%20the%201970s.
  3. JD Sports, “What is Flyknit Technology?” Accessed December 11, 2020.
  4. Sarah Scaturro, “(Im)mortal Fashion: Iris van Herpen’s ‘Skeleton’ Dress.” Accessed December 6, 2020.
  5. Iris Van Herpen, “Capriole.” Accessed December 6, 2020.
  6. Marketing Revolution, “What is AI Marketing?” Accessed December 5, 2020.,efforts%20where%20speed%20is%20essential.
  7. Join Marketing, “Fashion influencers and how they transform the fashion industry.” Accessed December 10, 2020.
  8. Saku New York, “Fashion Influencers.” Accessed December 11, 2020.
  9. How to Generate (Almost) Anything, “Human-AI Collaborated Fashion.” Accessed December 10, 2020.
  10. Wikipedia, “Generative adversarial network.” Accessed December 10, 2020.
  11. XR Goes Pop, “Acne Studios x Robbie Barrat.” Accessed December 6, 2020.
  12. Rob Dozier, “This Clothing Line Was Designed By AI.” Accessed December 6, 2020.
  13. Mark Liu, “The robot wears Prada: what happens when AI starts giving out fashion tips?” Accessed December 6, 2020.
  14. Will Knight, “Amazon Has Developed an AI Fashion Designer.” Accessed December 6, 2020.
  15. Katrina Lake, “Stitch Fix’s CEO on Selling Personal Style to the Mass Market.” Accessed December 6, 2020.