Integration of Artificial Intelligence in the Education of Future Teachers of Physics

The transformative potential of AI in education, is based on the architecture of transformers that are able to diagnose students' misconceptions with high accuracy and hence validity. Nevertheless, students still face persistent cognitive difficulties in learning physics, including misconceptions and gaps in background knowledge. Therefore, the hypothesis of the study was to hypothesize the feasibility of using ChatGPT as an additional pedagogical tool within a constructivist inquiry-based approach to improve conceptual understanding and address misconceptions in a thermodynamics course for freshman students who are future physics teachers. The purpose of the study was to examine the effectiveness of using ChatGPT, a generative AI chatbot, as an additional pedagogical tool in a thermodynamics course for freshmen students, future physics teachers. The scientific novelty of the study lies in revealing the mechanism of improving students' conceptual understanding of thermodynamics principles, eliminating misconceptions in the knowledge component and creating an interactive learning environment through ChatGPT, opening new opportunities for improving educational practices in professional education. The results of the study showed that the experimental group of students significantly outperformed the control group in improving conceptual understanding and reducing qualitative misconceptions. The results achieved emphasize the great potential of ChatGPT in STEM education within a constructivist learning environment to transform it. The practical significance of the research results is presented by a specific pedagogical toolkit that can act as an additional means of improving conceptual understanding and eliminating misconceptions in a thermodynamics course for freshman students - future physics teachers.