Professional Obsolescence Anxiety and the Use of Generative AI in Engineering Students: A Sequential Mediation Model

Abstract

Our study investigates the paradoxical relationship between professional obsolescence anxiety and the use of generative artificial intelligence (GAI) among engineering students, exploring the psychological mechanisms underlying why those who fear losing their jobs to technology are less likely to adopt AI tools. A sequential mediation model was tested using structural equation modeling (PLS-SEM) with 258 engineering students from Peru. The model examined the relationship between professional obsolescence anxiety, AI usage frequency, and the mediating factors of technological Self-Efficacy and performance expectations. Bootstrap confidence intervals were used to assess the indirect effects. Results revealed a significant sequential mediation pattern where professional obsolescence anxiety negatively affects technological Self-Efficacy (β = -0.116, p < .05), which then positively affects performance expectations (β = 0.525, p < .001), ultimately increasing AI use frequency (β = 0.609, p < .001). Notably, the study identified an inconsistent suppression phenomenon where direct effects (+0.077) and indirect effects (-0.068) work in opposite directions, resulting in no significant overall effect. The full mediation model showed a better fit compared to partial or no-mediation models. This research provides new insights into the technology adoption paradoxes by demonstrating that anxiety about professional obsolescence generates internal psychological conflicts that both encourage and hinder AI adoption. The findings have important implications for engineering education, indicating that interventions should focus on enhancing Technological Self-Efficacy rather than just providing access to AI tools. However, these findings should be interpreted cautiously due to the cross-sectional design and the single-institution sample, highlighting the need for replication in diverse educational contexts.

Ähnliche Arbeiten