Effectiveness of the AI using different typologies of design methods

Abstract Artificial Intelligence is emerging as a powerful tool within the field of product design. As such, its creative potential is the subject of numerous studies, as well as the advantages it provides to designers. In this context, the present study analyzes the use of different types of design methodologies by AI to generate creative product concepts. The objective of this study is to understand how designers can leverage AI as an assistant in the design process and help them select the most appropriate methodologies to work with AI according to their purpose. For this purpose, a GPT-4 based AI has been required to solve three problems with different degrees of difficulty using different typologies of design methodologies. The results show differences between the creativity values of the results when using different methodologies, as well as a difference when varying the complexity of the problem. The findings of the study may contribute in future for the development of AI assistants for designers or to develop AI-based tools to assist in conceptual design in a collaborative way.