Proceedings of the International scientific and practical conference “Science, Technology and Culture: Strategies for Sustainable Development” (December 15-17, 2025) / Publisher website: www.naukainfo.com. – Krakow, Poland, 2025. – 120 p.

76 misclassification poses a direct threat to principles of fairness in education and academic publishing. To realize the full potential of AILL while mitigating its risks, pedagogically grounded implementation is required-one that integrates AI-driven technological support with meaningful human involvement. This necessitates the adoption of a hybrid learning model that combines AI assistance with mandatory pedagogical guidance from instructors and critical evaluation of AI-generated outputs by learners. LLMs should be employed through structured prompting and under instructor guidance in order to prevent superficial learning. AI is most appropriately used for generating routine tasks (e.g., exercise creation and initial feedback), while instructors should focus on fostering students’ critical thinking, cultural awareness, and ethical use of these tools. REFERENCES: 1. Hockly, N. (2023). AI in Language Education: Opportunities and Challenges. ELT Journal, 77(4). Oxford University Press. 2. Large Language Model (LLM) (2024). Wikipedia. Available at: https://en.wikipedia.org/wiki/Large_language_model 3. Jiayu Huang. Empirical studies LLM Foreign Language Education post-2023 (2023). Springer. Available at: https://arxiv.org/html/2509.22725v1 4. Weixin, L. GPT detectors frequently misclassify non-native English writing as AI generated (2024). PMC. Available at: https://pmc.ncbi.nlm.nih.gov/articles/PMC10382961/