The role of ChatGPT and higher-order thinking skills (HOTS) as predictors of physics inquiry among upper-secondary students has yet to be widely explored. Therefore, this research aimed to examine upper-secondary students' role in ChatGPT (convenience and quality (CQ), motivation and engagement (ME), and accuracy and trust (AT)) and HOTS as predictors of physics inquiry. Data were collected from 334 upper-secondary students in Indonesia through online questionnaires and analyzed with SPSS software using correlation and multiple linear regression. The results showed that CQ had the strongest correlation with HOTS, with significant predictors being response speed, concept linkage, and explanation quality. The ME dimension was also significantly correlated with HOTS, with increased motivation to learn and enjoyment in learning as key predictors. Lastly, the AT dimension significantly correlated with HOTS, where the accuracy of information and students' trust in it were essential predictors. These findings indicate that ChatGPT has the potential to enhance inquiry-based learning in physics by effectively supporting the development of HOTS.
Keywords: physics inquiry, ChatGPT, higher-order thinking skills, correlation, multiple linear regression, AI in education