AI language models in an educational context: an analysis of communication strategies

Abstract

The article analyzes communication strategies for interacting with artificial-intelligence-based language models. The relevance of the topic stems from a scientific and didactic problem: how to ensure a controllable communication strategy with AI language models in educational settings that supports and develops learnersʼ cognitive skills and, in the long term, preserves the active role of a specialist in key stages of reasoning processes.

The aim of the study is to identify optimal strategies and rules for communicating with artificial-intelligence-based language models in order to obtain the highest-quality responses in the educational process, as well as to examine how user prompts influence the accuracy and relevance of generated outputs. Another important task is to develop clear guidelines that help learners acquire the ability to formulate precise, structured prompts and select effective communication strategies for working with language models.

The article proposes recommendations for choosing communication tactics that transform the use of AI language models into a fully fledged intellectual activity that preserves learnersʼ cognitive engagement, fosters the development of language competence, and supports adherence to the principles of academic integrity. Chain-of-thought and iterative prompting strategies, as well as single- and multi-step prompting methods, are identified as effective for the learning process, including for Ukrainian-language instruction. The study concludes that optimizing chatbot personalization settings significantly improves the accuracy and relevance of responses, while precise instruction-based prompts, the use of a professional style, and the specification of user parameters lead to substantially higher-quality outputs suitable for educational use. Outputs generated in Ukrainian require particular attention, and their verification should rely on authoritative dictionaries and the updated 2019 edition of Ukrainian orthography.

A well-substantiated approach to communication with artificial-intelligence-based language models, supported by clear prompts and specialized instructions in the context of language teaching and learning, will be useful for researchers and methodologists in preparing methodological materials and recommendations on the use of AI tools in education. A promising direction for further research is the development of a prompt repository for learnersʼ independent work in language study and a comprehensive evaluation of its effectiveness.