The advanced Gen AI tools have the capability to produce new text and imagery, paving the way for streamlined fashion design workflows.
The study was conducted by researchers at Pusan National University, South Korea, and the research team was led by professor Yoon Kyung Lee and Master’s student Chaehi Ryu from the Department of Clothing and Textiles.
Go deeper with GlobalData
This exploration underscores the need for advancements in AI to ensure its effective deployment in the realm of fashion design.
“To use AI effectively in fashion, we must understand the characteristics of generative AI models and make informed judgements of where they can be applied. In this study, we studied how effective prompt engineering can be used to generate realistic fashion collection images through AI,” professor Lee stated.
In their study, the researchers used ChatGPT-3.5 and ChatGPT-4 to analyse historical men’s fashion trends up to September 2021 and predict trends for the fall/winter season of 2024.
The research also focused on employing DALL-E 3 to capture seasonal fashion trends visually and to conceptualise novel fashion collections.
US Tariffs are shifting - will you react or anticipate?
Don’t let policy changes catch you off guard. Stay proactive with real-time data and expert analysis.
By GlobalDataThe team identified design elements from these predictions as “initial codes” and refined them with “modified codes” from Vogue’s trend data and “codes from literature” on fashion design concepts.
These were then consolidated into six final codes: trends, silhouette elements, materials, key items, garment details, and embellishments.
With these codes in hand, the team crafted 35 prompts for DALL-E 3 to visualise unique outfits for a hypothetical 2024 Fall/Winter fashion show.
Each prompt was executed three times, yielding a total of 105 images that incorporated various details such as aspect ratios, camera angles, model appearances, runway designs, and audience moods.
The AI model successfully interpreted the prompts accurately 67.6% of the time, particularly those with descriptive adjectives. Some generated images were found to closely resembled actual men’s fashion collections for the projected season.
However, challenges arose as DALL-E often defaulted to ready-to-wear styles and struggled with incorporating complex trend elements like gender fluidity, noted the study.
This suggested that while trend keywords are useful, they alone may not suffice for accurate generation without further model training.
“Our results show that expertly worded prompts are necessary for accurate fashion design implementation of generative AI, highlighting the important role of fashion experts. With further learning and improvements, generative AI models like DALL-E 3 will help fashion designers create entire fashion collections more efficiently, while supporting their creativity, and also help non-experts understand fashion trends,” professor Lee added.
The paper titled, ‘Effective Fashion Design Collection Implementation with Generative AI: ChatGPT and Dall-E’ was published in the Clothing and Textiles Research Journal on 22 June 2025.
Recently, Swedish fashion giant H&M unveiled its initial series of “digital twin” images of human models as part of its efforts to demonstrate the application of generative AI in enhancing creative endeavours in fashion.
Earlier this month, Spanish fashion retailer Mango also unveiled Mango Stylist, its new virtual fashion assistant using generative AI to customised product recommendations and enhance the shopping experience.
