Is employment anxiety among vocal music students associated with AI replacement concerns? The roles of AI anxiety and vocal-performance replacement perception.

Artificial intelligence (AI) is reshaping students' expectations about learning, employability, and professional identity, but AI-related psychological responses may differ across disciplines. This cross-sectional study examined how broad AI anxiety and a vocal-performance-specific replacement appraisal were associated with employment anxiety among 392 vocal music students and early-career graduates in China. AI anxiety was assessed with a 14-item subset derived from the Chinese Artificial Intelligence Anxiety Scale; vocal-performance replacement perception was assessed with a four-item context-adapted item set informed by the STARA awareness framework. AI anxiety was strongly associated with employment anxiety (r = 0.678, p < 0.001), whereas vocal-performance replacement perception showed a weaker zero-order association (r = 0.277, p < 0.001). Williams's test confirmed that the two dependent correlations differed, t (389) = 10.23, p < 0.001. In a hierarchical regression controlling for education stage/status, school type, and AI-use experience, AI anxiety had a positive coefficient (β = 0.589, p < 0.001), while vocal-performance replacement perception did not explain additional variance (β = 0.010, p = 0.824). Perceived career-direction clarity had a negative coefficient (β = -0.174, p < 0.001). Sensitivity analyses that removed the AI-anxiety work-replacement dimension or only its explicit replacement items yielded the same pattern. Within the set of AI-related variables examined, employment anxiety in this sample was more closely associated with broad AI anxiety than with an explicit belief that AI will replace vocal performance. The findings do not imply that replacement concerns are irrelevant or establish causal direction.