Impact of GenAI on ideation and innovation team performance
Description
This dataset accompanies the study titled "The Impact of Generative Artificial Intelligence on Ideation and the Performance of Innovation Teams" by Michael Gindert and Marvin Lutz Müller. The research investigates how a Generative AI (GenAI)-augmented ideation tool affects ideation quality, efficiency, and team dynamics within a structured innovation process. Utilizing the Knowledge Spillover Theory of Entrepreneurship (KSTE) as a framework, the data highlights the influence of AI on knowledge generation, transfer, and application during the ideation phase. A framed field experiment has been conducted. Two groups (experiment and control) had to ideate on two destinct innovation tasks. The experiment group used an AI augmented ideation tool, that has been developed for the study. Hypotheses: The dataset supports the testing of six main hypotheses concerning the application of large language models (LLMs) during ideation in team settings: H1: AI-augmented teams will produce higher-quality ideas due to enhanced knowledge spillover. H2: AI support will accelerate the ideation process. H3: Teams using AI will exhibit greater efficiency in idea generation. H4: AI-assisted teams will produce a more diverse set of ideas, in terms of novelty and feasibility. H5: AI-enhanced ideation increases the likelihood of generating revolutionary ideas. H6: AI-supported ideation will positively impact team satisfaction and engagement. Data Highlights and Findings: Idea Quality: The dataset includes quality assessments of ideas based on originality, feasibility, and clarity, revealing that AI-augmented teams consistently generated higher-quality ideas across various dimensions. Process Speed and Efficiency: Time measurements show that teams with AI support completed ideation tasks significantly faster than the control group, reducing ideation time by up to 30%. Diversity and Novelty: Evaluations indicate that ideas produced with AI support exhibited greater diversity in both approach and innovation potential, contributing to more revolutionary solutions, especially for complex problems in healthcare and automotive sectors. Team Satisfaction: Survey data suggest that team members in the AI-augmented condition reported higher levels of engagement and satisfaction, with a 23.3% increase in positive feedback compared to the control group.
Files
Steps to reproduce
A framed field experiment with n=70 has been conducted. Study participants have been randomized into control- and experiment group. within those groups the participants have been randomized in ideation teams of three people. All teams ideated on the same tasks: a healthcare product/service innovation for elderly and a business model innovation for automotive. All ideas have been blindly ranked according to the criteria for idea evaluation along Dean et al. (2010).