The Reskilling Paradox in the Age of Artificial Intelligence: Examining the Effects of AI Skills Training on Employees’ Job Security and Job Replacement Anxiety

Abstract

This study aimed to investigate the effect of structured artificial intelligence skills training on employees’ perceived job security and job replacement anxiety and to determine whether reskilling produces the paradoxical coexistence of greater occupational security and heightened awareness of AI-driven replacement risk. The study employed a quasi-experimental pretest–posttest design with an intervention group and a control group. The participants were 72 employees working in administrative, service, financial, and knowledge-based organizations in Tehran, Iran, who were assigned to an AI skills training group (n = 36) and a control group (n = 36). The intervention group participated in eight 90-minute AI skills training sessions over four weeks, whereas the control group continued routine occupational activities. Data were collected using the Job Insecurity Scale, the Artificial Intelligence Anxiety Scale with emphasis on the job replacement anxiety dimension, and an AI Skills and Training Competency Assessment. Data were analyzed using multivariate analysis of covariance and univariate analysis of covariance while controlling for baseline scores. The multivariate effect of group on the combined posttest outcomes was statistically significant, Pillai’s Trace = .731, F(3, 66) = 59.86, p < .001, partial η² = .731. After controlling for pretest scores, AI skills training significantly increased perceived job security, F(1, 69) = 49.72, p < .001, partial η² = .419, and AI skills competency, F(1, 69) = 224.54, p < .001, partial η² = .765. Job replacement anxiety also increased significantly in the intervention group compared with the control group, F(1, 69) = 11.53, p = .001, partial η² = .143. AI reskilling appears to produce a dual psychological effect by strengthening employees’ perceived occupational security and competence while simultaneously increasing awareness of AI-related replacement threats. These findings support the concept of a reskilling paradox and indicate that organizational AI training should combine technical skill development with transparent communication, career support, and strategies for managing replacement-related anxiety.