PROMPT ENGINEERING AS A TOOL FOR OPTIMIZING THE PROFESSIONAL ACTIVITIES OF EDUCATORS: AN EXPERIMENTAL STUDY OF THE EFFECTIVENESS OF GENERATIVE LANGUAGE MODELS

Abstract

Summary. Introduction. Modern education is undergoing a transformation driven by digital technologies and the impact of artificial intelligence (AI) on the learning process. One of the most innovative directions is the application of generative language models and prompt engineering to optimize routine educational tasks. Given the prevalence of clip-thinking and information overload among Generation Alpha, traditional learning approaches require adaptation. Prompt engineering offers new opportunities for automating educational content creation, personalizing learning processes, and improving overall efficiency.

Purpose. This study aims to develop a prompt creation methodology to optimize routine educational tasks and test it on various chatbots (ChatGPT, Copilot, Claude). The key objectives include analyzing current approaches to prompt engineering, developing a methodology for creating effective prompts, experimentally testing the proposed approaches, and evaluating the results.

Methods. The study is based on an analysis of scientific sources, the development of structured prompts, and their testing in different AI systems. Experimental analysis and statistical data processing methods (Kruskal-Wallis H-test) were used to assess the relevance, completeness, accuracy, practical applicability, and structural organization of chatbot-generated responses.

Results. An analysis of chatbot responses showed that different models have unique strengths in content generation. Claude demonstrates higher response relevance but lags behind ChatGPT in structural organization. Copilot exhibits stability and compliance with educational standards but sometimes requires additional adaptation. The study results confirm the effectiveness of the proposed prompt creation methodology, enabling educators to reduce lesson preparation time and tailor materials to students’ needs.

Originality. This study is innovative in the field of educational prompt engineering, as it not only analyzes the potential of generative AI but also proposes a clear methodology for creating prompts to optimize the educational process. Unlike previous studies, this work includes an experimental evaluation of different chatbots’ effectiveness and their comparative assessment.

Conclusion. The study results confirm that prompt engineering is a promising tool for automating routine educational tasks, personalizing learning content, and enhancing the efficiency of the educational process. Future research may focus on expanding the range of academic disciplines, adapting prompts for different age groups, and integrating AI technologies into digital educational platforms