[1G4-OS-13b] OS-13

4:10 PM - 4:25 PM JST(7:10 AM - 7:25 AM UTC)

[1G4-OS-13b-03] Effects of Collaboration with AI Trained on Others’ Thinking Styles on Outcomes of Creative Tasks

〇Kyosuke Hino¹, Hajime Sasaki^2,3 (1. DENTSU INC., 2. The University of Tokyo, 3. Dentsu Digital Inc.)

Keywords:

AI Persona,Personalized AI,Creativity,Sycophancy

Personalization in Large Language Models (LLMs) creates a risk of sycophancy (conforming to user preferences) and bias reinforcement. This can lead to narrowed perspectives and rigid thinking, which is particularly problematic in creative tasks where divergent thinking and unique viewpoints are essential. Based on the hypothesis that collaborating with an AI trained on others' thinking styles enhances creativity more than one trained on the user's own style, we conducted an experiment involving an advertising copywriting task. Participants collaborated with AI models trained on different thinking styles. Results showed that AI trained on others' styles facilitated a broader range of ideas, a sense of discovery, and higher originality. However, participants reported lower satisfaction with the final output, suggesting a trade-off between creativity and productivity. These findings highlight the importance of strategically incorporating "otherness" into the design of human-AI collaboration for creative work.