Product ideation in the age of artificial intelligence: insights on design process through shape coding social robots

Chakraborty, Shujoy; Loyens, Dirk; Aston, Jeremy

http://hdl.handle.net/10400.22/30681

Use this identifier to reference this record.

Name:	Description:	Size:	Format:
ART_DirkLoyens_2025.pdf		8.01 MB	Adobe PDF	Download

Send Feedback

Authors

Chakraborty, Shujoy

Loyens, Dirk

Aston, Jeremy

Abstract(s)

This research explores the impact of generative artificial intelligence (GenAI) on ideation and concept design of social robots capable of undertaking sustained long-duration human–robot interaction. The work reported here was developed between 2021 and 2024 through classroom teaching executed in four editions of 3-day project workshops involving 36 product design master students producing 27 concept design proposals in a European Higher Education Institute (HEI). The first two workshop editions used only classical methods utilising semantic moodboarding, sketching, virtual 3D modelling, and rendering. The last two editions employed mixed methods blending classical methods with computational methods using text-to-image and sketch-to-image GenAI tools, like Midjourney, DALL-E, and Vizcom. The findings suggest that using mixed methods, which co-creates by combining organic and synthetic creativity, enhances the concepts’ numeric quantity, although the concepts’ creative quality remains questionable. The advantage of the computationally enhanced mixed methods over the traditional classical methods is the greater potential to overcome creative blockages in novice designers with weak ideation skills. Increasing the volume of concept exploration increases the serendipitous probability of arriving at successful outcomes. This research is a case study of GenAI implementation in classroom teaching, highlighting its benefits and limitations for design courses in HEIs.