When does working with AI empower or entrap university faculty? A dualistic passion perspective on job crafting

Counts differ because each database indexes a different set of publications. We treat OpenAlex as the canonical count; Google Scholar is not shown (no API, and crawling it violates its ToS).

Introduction Generative AI is now widely used in teaching, research, and academic writing, leading university faculty to engage in more sustained human-AI collaboration. However, existing research has largely focused on tool-related outcomes, offering limited insight into the psychological experiences and academic work practices associated with such collaboration. Methods We conducted two survey-based studies using independent samples of Chinese university faculty. Study 1 ( n = 252) examined whether human-AI collaboration engagement was associated with harmonious and obsessive passion, and whether independent self-construal moderated these associations. Study 2 ( n = 363) replicated these relationships and further tested whether the two forms of passion were related to job crafting. Data were analyzed using partial least squares structural equation modeling. Results Across both studies, human-AI collaboration engagement was positively associated with both harmonious and obsessive passion. Independent self-construal strengthened the association between collaboration engagement and harmonious passion, but not the association with obsessive passion. Study 2 further showed that harmonious passion was positively related to job crafting, whereas obsessive passion was negatively related to job crafting. Discussion Human-AI collaboration may involve different psychological experiences among university faculty, depending on how such collaboration is internalized. Specifically, harmonious passion and greater job crafting reflect an empowerment-related pattern, whereas obsessive passion and reduced job crafting reflect a more constraining pattern of AI collaboration. These results point to the importance of motivational quality in explaining variation in academic work behavior beyond a tool-centered perspective.