In recent years, there has been an active introduction of artificial intelligence in education (AIEd). Recently, one of the most popular AI tools — ChatGPT — is an example of a generative AI technologies, which create new content (different types of text, image, audio, video) in response to a user’s request. On the example of ChatGPT, the article considers the possibilities of using AI in various spheres of educational activity and the ways of increasing its efficiency with the help of generative technologies. To structure the ways of interaction with ChatGPT during its integration into the educational process, the COST model, which describes the receipt and exchange of information in the learning process, is used. In order to utilize the capabilities of ChatGPT more fully and to compensate for its shortcomings, it is necessary to formulate appropriate queries, for example, using a concept tree. In addition, it is necessary to formulate queries in such a way that the concept under investigation is concretized as much as possible. As an example, the article presents fragments of the developed concept tree representing the composition of educational activity and interrelations between its elements, which can be the basis for making queries to a chatbot. The quality of the query response is determined by the completeness of the input information. The results of the research presented in the article will help to get more accurate and relevant answers to queries in the practical application of ChatGPT in the daily work of an educational institution by students and teachers as well as other stakeholders who are involved in the organization of the learning process.