Enterprise AI Analysis

The findings provide guidance for multiple stakeholders. Educators can embed AI-PLPs to deliver differentiated trajectories aligned with learners' proficiency, sustaining competence and engagement. Instructional designers should emphasize real-time feedback and scaffolded challenges that enhance both competence and engagement. Developers can integrate intuitive interfaces, and adaptive help functions to reduce dependence on external support. To fully utilize AI-PLPs for vocational and professional education, policymakers should invest in infrastructure as well as training.

Several limitations need to be acknowledged, providing directions for future research. One such limitation is the cross-sectional design, which restricts causal inference. Longitudinal or experimental studies would help examine how the competence-engagement-outcome pathway develops over time, which could further advance self-determination theory in AI learning contexts. Second, while the sample was drawn from vocational col-leges in Zhejiang Province, it covered multiple majors, providing reasonable diversity. Future research could further enhance gen-eralizability by including institutions from other regions. Third, the study was conducted within a single cultural context, which constrains the extent to which the findings can be generalized. Future studies could explore whether differences in autonomy and collectivism influence the effectiveness of AI-PLPs. Fourth, the measurement of technical support was based on self-reported data. Using analytics or log data could provide a more accurate reflection of actual usage. Finally, the current model had a limited scope. Incorporating perspectives from self-regulated learn-ing theory, flow theory, or cognitive load theory could lead to multi-theoretical models that better explain adaptive AI learning processes.