My Way or the AI Way? How Autonomy Frustration Undermines Creative Performance Through Editing AI Advice

ABSTRACT Artificial intelligence (AI) tools have become an integral part of organizational work, such as providing task‐relevant advice. Although existing research highlights the potential benefits of human‐AI collaboration for improving creative performance, recent findings question the universality of these effects. However, the mechanisms that may drive these differences in performance outcomes remain unclear. Drawing on self‐determination theory (SDT), we propose that receiving AI‐generated advice may undermine individuals' sense of autonomy, which is linked to a counterproductive process of individuals attempting to regain their autonomy and a reduction in their creative performance. We test this idea in a mixed‐methods design. In a first qualitative study, we find that people often experience a perceived loss of autonomy when working with an AI, which is associated with editing the AI's output. Building on these findings, in a second quantitative study, we find that autonomy frustration is linked to more extensive editing of the AI's output, which is associated with lower objective creative performance. Our findings underscore the importance of motivational factors, particularly the need for autonomy, for successful human‐AI interactions.