Annotation

The article examines the legal nature of the interaction between a learner and adaptive educational systems based on artificial intelligence within the framework of the constitutional doctrine of a human-centric state, as set forth in Decree of the President of the Russian Federation No. 309 of May 7, 2024. It demonstrates that the decisions made by such systems are quasi-administrative in nature — rendered without human involvement yet producing tangible educational consequences — while current legislation fails to account for this specific character. The key risks are analyzed: covert algorithmic discrimination, the psychological pressure of machine-generated recommendations, the erosion of the teacher’s professional role, and the practical impossibility of appeal. In response to these challenges, three interrelated regulatory mechanisms are proposed: the right to an explainable algorithmic decision, the imperative of human control, and a communication-law standard for interaction with an adaptive AI system. The article argues that implementing this triad makes it possible to integrate the technology into the legal framework as a «trusted information intermediary,» eliminating the existing information asymmetry and creating predictable conditions for all participants in the educational process.