Relative devaluation in AI transitions: an organizational behavior perspective on career sustainability beyond...

AI-enabled and digitally benchmarked evaluation systems are increasingly used in professional assessment, yet their implications for perceived professional value remain insufficiently understood. Drawing on Chinese higher education as a highly metricized institutional context, this study examines algorithmic relative devaluation (ARD), defined as professionals' perception that their contribution is comparatively downgraded when algorithmic or AI-mediated standards become institutionally privileged. Using a three-wave time-lagged survey of faculty members (N = 448) and PLS-SEM analysis, the findings indicate that algorithmic visibility and benchmark intensity are positively associated with ARD, whereas algorithmic explainability is negatively associated with ARD. ARD is, in turn, associated with both adaptive and defensive career reconfiguration and with lower perceived career sustainability. Adaptive career reconfiguration is positively associated with perceived career sustainability, whereas defensive career reconfiguration is negatively associated with perceived career sustainability. The findings further suggest that responsible AI governance plays an ambivalent role: it weakens the devaluing association of algorithmic visibility but strengthens the devaluing association of benchmark intensity, while not significantly moderating the explainability-ARD relationship. Rather than demonstrating causal restructuring of professional hierarchies, the study offers evidence that perceived AI-enabled evaluation features are linked to comparative worth perceptions and divergent career-related responses in a metric-intensive academic setting. The findings contribute to research on algorithmic management and sustainable careers by positioning ARD as a context-specific mechanism through which AI-mediated evaluation may shape perceived professional value beyond job insecurity and well-being concerns.