AI-Augmented Brainwriting: Investigating the use of LLMs in group ideation

AI-Augmented Brainwriting: Investigating the use of LLMs in group ideation

May 11–16, 2024 | Orit Shaer, Angelora Cooper, Osnat Mokryn, Andrew L. Kun, Hagit Ben Shoshan
This paper explores the integration of large language models (LLMs) into the creative process, specifically in group Brainwriting, a structured technique for generating ideas. The authors designed a collaborative group-AI Brainwriting framework that incorporates an LLM to enhance both the divergence and convergence stages of idea generation and evaluation. The study involved 16 students in an advanced undergraduate course on tangible interaction design, where they used an online whiteboard (Conceptboard) to generate ideas individually and then collaboratively with the help of an LLM. The LLM was used to generate additional ideas, which were then reviewed and discussed by the group. The evaluation of the framework included a user study and an LLM-based evaluation engine that assessed idea quality based on relevance, innovation, and insightfulness. The results suggest that integrating LLMs into Brainwriting can enhance both the ideation process and the outcome, with LLMs also supporting idea evaluation. The paper contributes to HCI education and practice by providing a new pedagogical framework and empirical insights into the challenges and opportunities of incorporating AI into collaborative ideation.This paper explores the integration of large language models (LLMs) into the creative process, specifically in group Brainwriting, a structured technique for generating ideas. The authors designed a collaborative group-AI Brainwriting framework that incorporates an LLM to enhance both the divergence and convergence stages of idea generation and evaluation. The study involved 16 students in an advanced undergraduate course on tangible interaction design, where they used an online whiteboard (Conceptboard) to generate ideas individually and then collaboratively with the help of an LLM. The LLM was used to generate additional ideas, which were then reviewed and discussed by the group. The evaluation of the framework included a user study and an LLM-based evaluation engine that assessed idea quality based on relevance, innovation, and insightfulness. The results suggest that integrating LLMs into Brainwriting can enhance both the ideation process and the outcome, with LLMs also supporting idea evaluation. The paper contributes to HCI education and practice by providing a new pedagogical framework and empirical insights into the challenges and opportunities of incorporating AI into collaborative ideation.
Reach us at info@study.space
[slides] AI-Augmented Brainwriting%3A Investigating the use of LLMs in group ideation | StudySpace