Technostress and Security Compliance in AI-Enabled Banking

This study examines how technostress is associated with employees’ information security compliance behavior among employees of a South Korean bank, an organization that has adopted AI-enabled information systems, and investigates the moderating role of perceived organizational support (POS). Drawing on Conservation of Resources theory and Organizational Support Theory, technostress is conceptualized as a multidimensional construct comprising techno-overload, techno-invasion, techno-complexity, techno-insecurity, and techno-uncertainty. Using survey data from 257 employees, hierarchical regression analysis shows that all five technostress dimensions are negatively associated with security compliance behavior. POS significantly moderates only the techno-overload–compliance relationship (corrected β = .044, p < .05); the four remaining interaction terms are not significant, and a nested-model test indicates that the moderating block as a whole does not reach conventional significance, ΔF(6, 242) = 2.03, p = .06. These results indicate that general technology-induced stress—measured with an established, general-purpose technostress scale rather than an AI-specific instrument—is associated with lower security compliance, and that organizational support offers only a partial, targeted buffer. This study contributes to the information systems literature by testing technostress as an antecedent of security compliance behavior in a high-security, AI-enabled organizational setting, while explicitly delineating the boundary between general technostress and AI-specific stress mechanisms (e.g. algorithmic opacity), which the present measures do not distinguish and which future research should operationalize directly.