More flexible or more rigid? A study on the double-edged sword effect of Gen AI usage dependence on employees’ emotional labor

Frontline service employees' emotional labor strategies (deep acting and surface acting) directly shape customer service experience value. As Generative AI (Gen AI) rapidly penetrates the service sector, human–AI collaboration is reshaping how employees perform emotional labor. However, the impact of Gen AI on emotional labor remains underexplored. Unlike prior research, we distinguish Gen AI usage dependence, specifically, innovative usage dependency and routine usage dependency, and link these to emotional labor. Drawing on conservation of resources (COR) theory, we developed an integrative research framework to examine whether and how Gen AI usage dependence affects emotional labor, with particular emphasis on the mediating role of cognitive flexibility and the moderating role of work stress (Challenge vs. Hindrance). Using a three-stage field survey, we analyzed data from 399 frontline service employees. The results indicate that (1) Gen AI usage dependence has a significant double-edged effect: innovative usage dependency strengthens cognitive resources and promotes deep acting, whereas routine usage dependency induces cognitive resource rigidity and leads to surface acting. (2) Cognitive flexibility mediates the relationship between Gen AI usage dependence and emotional labor. (3) Challenge-type work stress strengthens the positive effect of innovative usage dependency, while hindrance-type work stress amplifies the negative effect of routine usage dependency. In addition to extending theory on Gen AI and emotional labor, this study suggests that managers should strike a dynamic balance between promoting Gen AI adoption and optimizing employees’ emotional labor strategies to build a sustainable human–AI collaboration environment.