Creativity + AI · Support Tools

Study: Problem Reframing

Problem reframing is a crucial designerly activity that can make a challenging design problem more solvable. However, problem reframing is inherently challenging and labor-intensive, even for expert designers. It demands putting great effort into understanding problems and altering perspectives.

We look at whether working with AI–specifically, large language models (LLMs)–benefits designers in devising high-quality problem frames. We observed how expert and novice designers (N = 280) generate alternative ways of reframing design problems without LLM and three conventional ways of using LLMs (i.e., free-form, direct generation, and a structured approach informed by Dorst's theory of reframing). From this, we contribute novel findings on understanding the influence of using LLMs.

Study: Personas

Personas are fictional characters created from user data to represent potential users of a product or service. Personas personify user data to encourage perspective-taking and evoke empathy toward potential users of a product or service. However generating high-quality representative and empathy-evoking personas is challenging and laborious.

We studied whether designers and large language models (LLMs) can create personas from qualitative user data together. Our findings suggest that the most representative and empathy-evoking personas are created when designers group user data by their key characteristics, while LLMs summarize the pre-grouped user data and generate the narratives for the personas.

Study: Chatbots in Co-Design

Consensus-building is an essential process for the success of co-design projects. We ask whether conversational agents such as chatbots could facilitate asynchronous consensus-building in large co-design projects with multiple stakeholders.

Our results indicate that interacting with a chatbot can increase stakeholders’ willingness to commit to group decisions by eliciting the perceived joint effort and fairness, also in the face of group decisions that conflict with one’s own opinions. We discuss design implications for facilitating consensus-building with chatbots and the potential benefits that chatbots may bring to asynchronous co-design.

Workshops