Perceived control influences users' acceptance of AI-enabled services

With the rapid diffusion of artificial intelligence (AI) in the service sector, users' acceptance of AI-enabled services has become an important issue for successful deployment. Drawing on control-related and appraisal-based perspectives, the present research examined whether perceived control is associated with acceptance intention toward AI-enabled services, whether AI anxiety may represent one emotional pathway underlying this relationship, and whether robot role may serve as a boundary condition. Across three scenario-based experiments with college student participants (Study 1: N = 190; Study 2: N = 280; Study 3: N = 360), Study 1 showed that higher perceived control was associated with stronger acceptance intention. Study 2 replicated this relationship and found that AI anxiety partially accounted for the relationship between perceived control and acceptance intention. Study 3 provided preliminary evidence that robot role may moderate this relationship, with the association between perceived control and acceptance intention appearing relatively stronger in the companion-type robot condition than in the tutor-type robot condition. These findings suggest that perceived control may be a relevant psychological factor in AI-enabled service acceptance and highlight the roles of users' emotional responses and role expectations in AI-enabled service interactions.