Conference Agenda

The problem-solution co-evolution model is a foundational framework for under-standing the emergence of creativity in both individuals and teams. With the ad-vent of Generative Artificial Intelligence (GAI), a new paradigm of co-creation in conceptual design has arisen. However, the dynamics inherent to human-GAI col-laborations remain largely unknown. In our investigation of the co-evolution dy-namics of human-GAI interaction, we employed retrospective protocol analysis to examine the verbal reasoning processes of twenty novice designers co-designing with GAI (text-to-text and text-to-image models). Drawing from the outcomes of our creativity assessments, a key revelation emerged: GAI has the potential to amplify team creativity by fostering human abductive reasoning. In further discourse, we introduce a novel human-GAI co-evolution model, which elucidates the significant role of GAI in aiding human problem-framing explora-tion. Central to our exploration, we spotlight "introspection" and "retrospection" as pivotal constructs in probing human-GAI collaborations.

Sketching is quick and effective, however with the advent of generative AI, do the current generation of novice designers have an alternative? This paper compares the use of sketching and text-to-image generative AI tools to produce initial concept images (“sketches”) by novice designers. This will identify the viability and potential adoption of AI as a replacement, and gauge the adoption willingness of novice designers, replacing sketching. This study compares conventional sketching and AI image generation using first year product design students to record brainstormed initial concepts using both sketching and generative AI tools, this study compares various attributes of both, including ability to represent designer’s intentions. The findings of this study suggest that at present, novice designers continue to prefer conventional sketching with 75% believing that it is more accurate to designers’ intentions and 59.62% believing that it is easier to use.

Artificial Intelligence Generated Content (AIGC) introduces a new collaborative design paradigm where words, sentences, and images circulate within the team as new design knowledge. However, due to the limited controllability and inter-pretability of current AIGC models, collaboration between designers and AI de-mands continuous iterations and experimentation. How to establish a reusable representation for the knowledge of the collaborative process is an open prob-lem. Our comprehensive approach, including focused interviews, case studies, and workshops, revealed transmission patterns of design concepts during both divergent and convergent phases. To represent the interaction between design-ers, we propose a novel node-based design strategy, where each node is an AI operation with its prompts and outputs and each link denotes the data flow to the next node. Implementing this strategy, we crafted a design system that en-hances synergy between the design team and AIGC.

The emergence of generative AI sparked thoughts on how it can be helpful across the stages of fashion design practice and creativity. We investigated the potential impact of prompt engineering using stable diffusion and Midjourney on understanding the relationship between the prompt and outcome, and how it influences the final process, thinking, or result. Our study provides guidelines for designers to use specific formats and words to describe the garments to generate clothing designs even if they are not familiar with text prompt engineering. We found fashion designers more likely to focus on the clothes and the overall feeling that the images evoke, meaning that anatomical accuracy is not their concern. This research is a rare exploration that combines prompts engineering with fashion, providing insights and recommendations on how to better utilize AI as a tool in industry.

The summer of 2022 marked the advent of accessible text-to-image tools, revolutionizing image rendering with distinctive styles swiftly. This generated a creative shift among designers, generally addressed as “prompt design”, although this expression scarcely captures the profound interaction between design and digital tools. This paper elucidates the potential synergy between designers and AI through two pragmatic exercises engaged by university students. Our approaches were polarized: in one exercise we fostered a rich imaginative process before the text-to-image creation; secondly we asked students to elaborate a possible user interface over an artifact drawn by AI, following a very simple textual description.

The result is a framework that combines both the relevance of structured imaginative process and the capabilities of generative AI technologies, supporting an enriched dialogic interaction between design and dataset-based imagery.

In the rapidly advancing landscape of generative AI, the role of designers is constantly being reshaped amidst the choice between traditional tools and emerging technologies. To comprehend this transformation, we developed a platform integrating multiple generative AI models and allowed for unrestricted search and sketching to emulate a realistic working scenario. Within this setting, we analyzed the design behavioral patterns exhibited by experienced designers: through both quantitative and qualitative analyses, we examined the cause and effect of generation patterns and iteration patterns. This exploration inspired insights into the designer's evolving role, transitioning to a continuous learner and being more than a curator when incorporating generative AI.

Digital narrative has great potential in the field of cultural relics. This visual paper proposes an AI-based digital narrative design method for cultural relics and visualizes their contexts. Based on the study, this method constructed the narrative by taking cultural relics as the main body and referencing historical materials such as literature, ancient paintings, film, and television works. Midjourney was used to generate contextual images, and digital narrative scenes were constructed using mental canvas. Added sketch notes in post-synthesis to complete the work. Finally, in the design practice of "The Silk Road around Chang'an", feedback was collected through the Likert scale to verify the effectiveness of visual narratives and sketch notes in understanding cultural relics. Overall, this study redefines the cooperative relationship between designers and AI in the digital age and provides new perspectives and methods for digital humanities in the future.