Online searching tools are used for early design ideation. However, they tend to limit designers’ creativity due to direct visual feedback and lack of novelty. We present IdeaSquares, a design concept text generation tool for supporting the ideation process. We used text-generation artificial intelligence (AI) that was fine-tuned on the Red Dot design award winners’ data to show newly generated design examples as search results. To understand how young designers use concept text generation tool for the design ideation, we conducted a comparative user study where participants used two versions, generative querying and conventional searching mode, of the tools during the ideation task. Our findings revealed that (1) texts are combined and reinterpreted to different meanings during the ideation and (2) ideas are developed by finding appropriate keywords that fit to a primitive idea when the generative querying mode was used. Based on the tool development and the study results, we discuss how generative texts empowered by AI can be used as a source of inspiration and further support the design ideation. 
Yun, G., Cho, K., Jeong, Y., and Nam, T. (2022) Ideasquares: Utilizing generative text as a source of design inspiration, in Lockton, D., Lenzi, S., Hekkert, P., Oak, A., Sádaba, J., Lloyd, P. (eds.), DRS2022: Bilbao, 25 June - 3 July, Bilbao, Spain. https://doi.org/10.21606/drs.2022.484

윤경원, 남택진, "디자이너의 창의적 사고를 돕기 위한 텍스트 기반 인공지능 디자인 생성 도구", HCI KOREA 2021, 온라인, KO, 2021.01.27
Back to Top