THE POSSIBILITIES OF USING ARTIFICIAL INTELLIGENCE TO IMPROVE THE EFFECTIVENESS OF CHEMISTRY LESSONS

Abstract

The article examines the potential of artificial intelligence (AI) tools—virtual laboratories, generative neural networks, and adaptive learning platforms—for improving the quality of chemistry teaching in secondary schools in Kazakhstan. In a quasi-experimental study, 42 ninth-grade students from the A.Navoi School No. 13 (Shymkent) were randomly assigned to an experimental group (AI-integrated program, n = 21) and a control group (traditional instruction, n = 21). Before and after the six-week course, students completed a 40-item test aligned with the state curriculum standards and a learning-motivation questionnaire. Following the course completion, the experimental group demonstrated test results 25% higher than those of the control group (d = 0.82; p < 0.001) and a higher engagement level (85% vs. 55%). A two-way ANOVA revealed a significant Group × Time interaction (F(1, 38) = 12.54; η² = 0.21). Additionally, a 30% increase in learning independence and more substantive discussions of organic chemistry topics were observed. The findings indicate that AI enables safe and resource-efficient experimentation, individualizes the learning pace, and reduces students’ cognitive load. Methodological recommendations—including 30–45-minute VR laboratory sessions, generative reaction visualization, adaptive diagnostics, estimated implementation costs, and a teacher-training model for large-scale adoption—are proposed. Further longitudinal research is required to evaluate long-term effects and implementation in rural schools.