Generative Adversarial Networks
Another AI-focused trend that began developing in the fashion industry pre-pandemic is generative adversarial networks (GANs), a highly representative form of generative AI technology used to model data distributions. GANs are a type of auto-generative modeling that uses machine learning to automatically discover and learn the regularities or patterns of data to generate or output new examples that plausibly could have been drawn from the original dataset (Brownlee, 2019). This means GANs can generate unique patterns, images, and other creative designs that was thought to be only a human trait.
Some current examples of GANs being utilized in the fashion industry are: TextureGAN creates multiple colors or patterns that are then applied to basic sketch images; multi-label AC-GA generates images of fashion products based on characteristics input as text; and CAGAN enables users to generate upper body and garment images (Sohn et al., 2020).
GANs can be used as powerful marketing and communications tool for younger generations of consumers. Members of Generation Z tend to be tech savvy and fashion-sensitive, with knowledge of fashion trends, which means GAN-generated fashion products may influence the perceptions and purchase intentions of tech-savvy members of Generation Z. This can also increase consumer engagement and purchasing since Sohn et al. (2020) suggests that GAN affects willingness to interact and pay for fashion since it allows individuals to apply their preferred images to fashion designs.