Emotional Fatigue or Support as Dual Pathways of AI Interaction on Worker Well-being in Smart Work Environments

Artificial intelligence (AI) systems increasingly mediate work processes, making employee communication, decision-making support, and task automation more seamless. However, the psychological implications of these technologies have become critical to understanding organizational roles and sustainability. This study applies the Job Demands–Resources (JD-R) Model to assess whether AI functions as a work resource that promotes motivation, reduces stress, and provides emotional relief through efficiency and support systems, or as a demanding agent that increases cognitive load, alienation, surveillance pressure, and emotional exhaustion. Relevant literature published between 2015 and 2025 was systematically sourced from Web of Science, Scopus, IEEE Xplore, PubMed, and Google Scholar using predefined search, screening, and exclusion criteria. Findings indicate that supportive AI, particularly in decision assistance, intelligent feedback, and automated task reduction, enhances employee well-being by reducing emotional strain and improving perceived competence. In contrast, AI systems that lack human-centered design, intensify monitoring, or increase work complexity tend to trigger emotional fatigue, anxiety, and reduced job satisfaction. The study concludes by emphasizing the need for human-centered AI design that balances efficiency with empathy, ensuring the protection of workers’ emotional well-being in technologically advanced workplaces.