How does generative AI learning support affect vocational college students’ employability? – An empirical analysis based on latent profiles and interdisciplinary learning chain pathways

In the era of artificial intelligence, the cultivation of vocational college students’ employability urgently needs to break through traditional skill frameworks to effectively respond to ongoing occupational transformations. Grounded in social cognitive theory and the technology acceptance model, this study employed latent profile analysis (LPA) to classify 2,622 vocational college students based on their multidimensional characteristics of Generative AI usage. Three distinct profiles were identified: rationality-dominant (62.9%), passively detached (24.7%), and balanced development (12.4%). A chain mediation model revealed a significant mediating effect of interdisciplinary learning motivation and interdisciplinary learning ability on the relationship between Generative AI learning support and employability (total effect β = 0.13, p < 0.001). Furthermore, for all three profiles, the association between generative AI learning support and employability was fully mediated by interdisciplinary learning motivation and interdisciplinary learning ability, as no significant direct effects were observed. Based on these findings, learners and educational administrators should focus on students’ cognitive development to avoid the “selective attention blindness phenomenon.” Simultaneously, a multidimensional framework for employability cultivation should be established, emphasizing the enhancement of students’ learning motivation and abilities to facilitate positive transitions in their Generative AI usage patterns.