IMPACT OF ARTIFICIAL INTELLIGENCE ON THE EDUCATIONAL ENVIRONMENT AND RISKS

Abstract

This article examines how artificial intelligence (AI) is reshaping the educational environment in Uzbekistan, with particular attention to two intertwined dimensions: pedagogical effectiveness and information security risks. The study conceptualizes AI in education as a socio-technical system that influences teaching practice, assessment, institutional management, and student learning trajectories through automated feedback, adaptive content delivery, generative tutoring, and analytics-driven decision-making. The article aims to identify the main instructional gains associated with AI adoption in pedagogical universities and related teacher-training contexts, while simultaneously mapping the risk landscape created by increased data collection, platform dependence, and the rapid diffusion of generative tools. Methodologically, the research synthesizes policy and institutional practice analysis with a structured review of recent empirical findings on AI-supported learning, and applies a risk-oriented framework to interpret challenges relevant to Uzbekistan’s digital education agenda. The results indicate that AI integration can improve personalization, formative assessment quality, and learner engagement when used to scaffold metacognitive strategies, provide timely diagnostic feedback, and reduce routine teacher workload. In addition, AI-assisted content generation can broaden access to instructional materials and support differentiated instruction for diverse student groups, including learners with varying language proficiency. However, the study also finds that these benefits are conditional on governance quality and pedagogical design. Information security risks emerge as a critical constraint: student data privacy vulnerabilities, insecure third-party services, weak consent practices, and insufficient data minimization can expose institutions and learners to leakage, profiling, and unauthorized secondary use. Generative AI introduces further threats to academic integrity through contract cheating, synthetic plagiarism, and assessment distortion, as well as to cognitive security via misinformation, hallucinated references, and overreliance that may reduce independent reasoning. Algorithmic bias and unequal access amplify risks of educational stratification, particularly when models trained on external corpora do not reflect local linguistic and cultural contexts. The article concludes that Uzbekistan’s teacher education sector requires a balanced adoption strategy that pairs instructional innovation with enforceable safeguards: privacy-by-design, secure procurement standards, integrity-preserving assessment redesign, AI literacy for faculty and students, and institutional monitoring mechanisms. The proposed synthesis provides a practical evidence-informed basis for universities to maximize pedagogical value while reducing exposure to information security and ethical hazards.