CHALLENGES OF INTEGRATING GENERATIVE AI INTO RUSSIAN LANGUAGE TEACHING IN KAZAKHSTANI UNIVERSITIES

Abstract

This article is devoted to studying the ability of Russian language instructors at higher educational institutions in Kazakhstan to identify academic texts generated by artificial intelligence. Experimental data show that even experienced specialists face difficulties in reliably determining text authorship, which demonstrates the high level of stylistic adaptation of generative AI to the conventions of academic writing. The study analyzes the gap between instructors' subjective confidence and the actual accuracy of detection, and also compares the obtained results with data from international research. Particular attention is paid to a comprehensive analysis of the pedagogical and ethical risks associated with the uncontrolled use of AI-generated texts in the educational process. The work substantiates the pressing necessity of integrating digital and "AI literacy" elements into curricula, fundamentally revising traditional assessment criteria, and strengthening academic integrity policies. The particular relevance of adapting Kazakhstan's linguistic education system to new challenges is emphasized, which requires developing students' critical thinking, reflection, and unique skills that cannot be replicated by artificial intelligence. The obtained results indicate the need to transition from reproductive tasks to creative ones, which will help maintain assessment objectivity and the integrity of the educational process.