Artificial intelligence in the fashion industry: consumer responses to generative adversarial network (GAN) technology

Abstract

PurposeThis study examines consumers' evaluations of product consumption values, purchase intentions and willingness to pay for fashion products designed using generative adversarial network (GAN), an artificial intelligence technology. This research investigates differences between consumers' evaluations of a GAN-generated product and a non-GAN-generated product and tests whether disclosing the use of GAN technology affects consumers' evaluations.Design/methodology/approachSample products were developed as experimental stimuli using cycleGAN. Data were collected from 163 members of Generation Y. Participants were assigned to one of the three experimental conditions (i.e. non-GAN-generated images, GAN-generated images with disclosure and GAN-generated images without disclosure). Regression analysis and ANOVA were used to test the hypotheses.FindingsFunctional, social and epistemic consumption values positively affect willingness to pay in the GAN-generated products. Relative to non-GAN-generated products, willingness to pay is significantly higher for GAN-generated products. Moreover, evaluations of functional value, emotional value and willingness to pay are highest when GAN technology is used, but not disclosed.Originality/valueThis study evaluates the utility of GANs from consumers' perspective based on the perceived value of GAN-generated product designs. Findings have practical implications for firms that are considering using GANs to develop products for the retail fashion market.

Values that drive desire: examining determinants of mobile promotion satisfaction

“…, 2021). For example (Sohn et al. , 2020) applied the five value dimensions of values – functional, social, emotional, epistemic and conditional – to the study of generative adversarial network technology in fashion retailing.…”

“…CVT's five dimensions can be further conceptualized into specific forms to incorporate consumption value in diverse contexts involving food consumption (Choe and Kim, 2018), streaming apps (Oyedele and Simpson, 2018), food delivery apps (Kaur et al, 2021) and mobile app usage (Zolkepli et al, 2021). For example (Sohn et al, 2020) applied the five value dimensions of valuesfunctional, social, emotional, epistemic and conditionalto the study of generative adversarial network technology in fashion retailing. Slack et al (2020) demonstrated that functional value (value for money and performance quality), social value (self-image and social approval) and emotional value drive satisfaction during shopping.…”

Big Data driven Product Design: A Survey

“…GAN and its extended models are unstable for design scheme generation, with low image quality and strong randomness. However, Sohn et al [226] demonstrated that artificial intelligence and GAN enhanced customer satisfaction and freshness. Their results showed the potential of GAN in product design.…”

Environmental Biology and Conservation Genetics Research in the Context of Artificial Intelligence

“…(1) Glue preparation: Mix 10ml of 50xTAE solution with 490ml of ultrapure water, mix evenly, configure into 1xTAE solution, then weigh 31.5g of urea into a beaker, and mix quickly with a glass rod; it can be taken out several times during the period Shake and cool to 50-60°C (the outer bottle body can be rinsed with cold water to achieve rapid cooling), add 3 μL of non-toxic nucleic acid dye, mix well; place the beaker in a water bath, heat at 65°C, and then quickly add 37.5 μL TEMED and 250 μL of 10% ammonium persulfate can be quickly stirred with a glass rod for more than ten seconds (the bottle mouth should not face people during stirring to prevent the liquid from splashing) [17][18].…”