GPT-4 demonstrates promising capability to automate the creation of personality situational judgment tests (PSJTs) in Chinese, offering a scalable alternative to traditional labor-intensive methods.

• Reduces development time and resources for personality assessments
• Addresses bias concerns in traditional test development
• Creates consistent, standardized psychological evaluation materials
• Enables on-demand generation of customized assessment scenarios

This innovation has significant implications for medical and clinical psychology settings, where efficient, standardized personality assessment tools can improve diagnostic processes, treatment planning, and mental health evaluations at scale.

Automatic Item Generation for Personality Situational Judgment Tests with Large Language Models