When AI feels unfair: The role of emotional depletion and AI literacy in shaping workplace deviant behavior

This study examines how perceived algorithmic injustice influences employees’ workplace deviant behavior. Drawing on Affective Events Theory, the study further investigates the mediating role of emotional depletion and the moderating role of AI literacy in shaping this relationship.Data were collected using a time-lagged survey design from employees working in knowledge-intensive organizations in major Canadian cities. A total of 398 questionnaires were distributed, resulting in 362 initial responses. After matching responses across two waves and removing unmatched cases and outliers, the final sample consisted of 335 employees. Structural equation modeling was employed to test the hypothesized relationships, including mediation and moderation effects. The results indicate that perceived algorithmic injustice significantly increases workplace deviant behavior both directly and indirectly through emotional depletion. Emotional depletion partially mediates this relationship, highlighting emotional strain as a key mechanism linking unfair AI-driven decisions to negative employee behaviors. Furthermore, AI literacy moderates the relationship between algorithmic injustice and emotional depletion, such that the relationship is weaker among employees with higher levels of AI literacy. The findings underscore the importance of fair and transparent AI systems in organizations. Managers should invest in enhancing employees’ AI literacy through targeted training programs to reduce emotional strain and mitigate deviant behaviors associated with perceived algorithmic injustice. This study extends the literature on workplace deviance by introducing algorithmic injustice as a technology-driven source of perceived unfairness in AI-enabled work environments. It contributes by identifying emotional depletion as a key mediating mechanism and demonstrating the moderating role of AI literacy as a critical individual capability. By integrating both affective and capability-based perspectives, the study provides a more comprehensive explanation of how employees respond to AI-driven decisions.