Implementation of Artificial Intelligence in an Online Platform for Solving and Checking the Solutions of Physical Problems

Solving physical problems plays a key role in the educational process, contributing to the achievement of a number of important goals: demonstrates the practical application of theoretical knowledge, develops logical thinking, concretizes acquired knowledge, connects theory with practice, and develops personal qualities of students. The object of the study is the process of learning physics using an online platform for solving (OPS) physical problems (PP). The subject of research is methods and means of using artificial intelligence (AI) for automated verification and evaluation of PP solutions on an online platform. The task of the research: to develop a front-end for independent solution of PP with the possibility of applying AI at different steps of the solution as a hint explaining the course of the solution, and a back-end with the implementation of modern AI models such as ChatGPT, Gemini, Copilot, Claude, for automated checking and assessments PP solutions. The obtained results include substantiating the advantages of using AI, including automation of the verification process, high accuracy of analysis, error detection, objectivity of evaluation, scalability, adaptation to user needs. The pedagogical consequences of the introduction of AI in physics education and the importance of a balanced approach for the development of critical thinking, creativity and consideration of ethical considerations are considered.