Automation or augmentation? The impact of artificial intelligence's technological characteristics on usage intention...

Artificial intelligence (AI) is revolutionizing clinical practice. As medical AI becomes increasingly a part of regular practice and procedures, a deeper understanding of the AI usage intention of medical staff is of great value. By incorporating technology-task fit and self-determination theory, this study developed a conceptual model through which to identify and observe the effects of medical AI automation and augmentation on the psychological needs (i.e., autonomy, relatedness, and competence), technology-task fit, and AI usage intention of medical staff. Using cross-sectional data from 400 Chinese medical staff, a partial least squares structural equation model (PLS-SEM) analysis showed that medical AI automation correlated positively with perceived autonomy and competence, while augmentation correlated positively with perceived autonomy, competence, and relatedness. Mediation analysis indicated that perceived autonomy, competence, and technology-task fit were sequential mediators in the association between medical AI automation and usage intention; perceived autonomy, competence, relatedness, and technology-task fit were sequential mediators in the link between medical AI augmentation and usage intention. These findings exemplify the automation-augmentation paradox in AI research, offering insight into the AI usage intention of medical staff and the underlying psychological mechanisms involved. Strategies to facilitate medical staff's clinical AI usage intention stemming from a human-centered perspective are also proposed.