THE USE OF LARGE LANGUAGE MODELS IN THE EDUCATIONAL PROCESS: AN EMPIRICAL ANALYSIS AMONG UNIVERSITY AND SECONDARY SCHOOL STUDENTS

Abstract

The increasing use of large language models (LLMs) in education has raised important questions regarding their impact on learning, trust, ethics, and data protection. This study examines the use of LLMs among secondary school and university students in Slovenia based on an online survey conducted in 2024 (N = 120). The main objectives were to analyze patterns of use, levels of trust and verification, perceptions of ethical issues, and the perceived impact of LLMs on learning efficiency. The study also investigates whether differences exist between educational levels and gender. The results indicate that LLMs are widely used for educational purposes, particularly for preparing assignments and presentations. Students report moderate trust in AI-generated information and frequently verify its accuracy, suggesting a cautious approach. Perceived time savings are significant, while privacy concerns remain high. No statistically significant differences were found between university and secondary school students or between genders. The findings highlight the growing integration of LLMs into education and emphasize the need for clear guidelines on their ethical and responsible use.