Validation of the Problematic AI Use Scale for University Students (PAIU-U) for Academic Integrity

Objective: This study reports the development and psychometric validation of the Problematic and (Adaptive) AI Use Scale for University Students. The scale’s theoretical foundation integrates behavioral addiction frameworks, AI-related ethical considerations, and self-regulated learning perspectives. Method: A total of N=1114 university students were surveyed online. Participants responded to the 27 PAIU-U items and additional validation measures. The sample was randomly split for calibration (n₁=543) and validation (n₂=571). An exploratory factor analysis (EFA) was conducted to identify the factor structure, followed by confirmatory factor analysis (CFA) on the second subsample. Results: EFA revealed a coherent five-factor solution accounting for 57.5% of variance, with high item loadings that mapped onto the theorized dimensions. The factors identified were: (1) Salience/Preoccupation & Tolerance–Escalation, (2) Functional Reliance/Capability Erosion, (3) Social and Ethical Conflict & Harm, (4) Loss of Control & Academic Impairment, and (5) Mood Modification/Coping. Conclusion: The PAIU-U is introduced as a novel measure of university students’ problematic AI use and associated academic integrity lapses. The scale demonstrates a robust multidimensional structure aligned with behavioral addiction theory and academic ethics and shows strong reliability and initial evidence of validity.