AI-enabled job demands and employee well-being: Emotional exhaustion and turnover across organizational contexts

As artificial intelligence (AI) becomes embedded in organizational work systems, employees' psychological responses increasingly shape how AI is experienced, enacted, and sustained in practice. This study examines how AI-enabled job demands influence employee well-being and turnover intention, and how these processes vary across organizational contexts (public and private sectors). Drawing on the Job Demands-Resources (JD-R) framework, we conceptualise AI job complexity and AI awareness as psychologically salient demands that condition employees' emotional and behavioral responses to AI integration. Drawing on time-lagged survey data collected from 456 employees using AI across public and private organizations, this study tests a health impairment process linking AI-enabled job demands to turnover intention via emotional exhaustion using PLS-SEM and multi-group analysis (MGA). The findings show that both AI job complexity and AI awareness increase emotional exhaustion, which in turn mediates their effects on turnover intention. AI awareness also directly predicts turnover intention, highlighting its role as a key psychological trigger of withdrawal under AI-driven work transformation. Sectoral comparisons further reveal that the emotional impact of AI awareness is significantly weaker in public organizations, while AI self-efficacy buffers the relationship between AI awareness and turnover intention only in the public sector. By foregrounding emotional exhaustion as a core psychological mechanism, this study advances understanding of how AI reshapes employee well-being, resistance, and retention. It contributes to interdisciplinary debates on human-AI interaction by demonstrating that AI's psychological consequences are contingent on organizational context and personal resources, offering actionable insights for designing human-centred, psychologically sustainable AI-enabled workplaces.