Evidence-based and fair assessment of students' competencies in the context of the availability of artificial intelligence

The widespread availability and ease of using artificial intelligence tools in the educational process of students lead to the issue of correctly assessing students' knowledge. The final test results do not always reflect the level of understanding and formation of students' competencies. The research purpose is to develop and evaluate the effectiveness of the implementation of a two-component assessment framework that combines assessment based on a number of criteria for written work and an interactive check of understanding of the written work. The potential of the proposed structure is being evaluated to increase the validity and fairness of student knowledge assessment in the era of artificial intelligence. The interactive oral test ensures the validity of the assessment of the original (written) assignment. Participants of the conducted empirical study were 2nd year students in “Pedagogical Education” at the discipline “Modern Educational Technologies” (N = 26, Tambov, 2025). The certification task included the development of pedagogical technology for elementary school students (free form). The comprehension test was carried out in an interactive format (the student formed an individual case by selecting 5 questions from different sections of the course, and answered them on the test). To quantify the identified combinations based on the results of a written assignment and an oral response, the study proposed and calculated an evidentiary results index (ERI). High level of evidence ERI ≥ 0.80, average level 0.60 ≤ ERI < 0.80, low level ERI < 0.60. It is proved that the use of a two-stage competence check involving the students themselves confirms the depth of knowledge gained and reduces the risk of falsely high results in written work, which is especially important in the AI environment.