An Adaptive Intelligent Tutoring System Powered by Generative AI
Habiba Almetnawy, Ahed Orabi, Alreem Rashed Alneyadi, Tasneim Ahmed, Abderrahmane Lakas · 2025
The emerging technology of Generative AI (GenAI) and its applications in various fields have opened new possibilities for educational technology, particularly in the development of advanced Intelligent Tutoring Systems (ITSs) as there is an increasing need for systems that can provide personalized learning experiences. Despite significant advancements in the development of ITSs, many proposed solutions struggle to effectively adapt to diverse learner profiles, creating barriers to personalized and adaptive learning experiences. This paper explores the transformative potential of GenAI in developing ITSs and proposes a novel ITS powered by a Generative Pre-trained Transformer (GPT), which leverages advanced Large Language Models (LLMs) to deliver contextually relevant tutoring sessions tailored to individual learning styles and proficiency levels. Key aspects of the methodology used in this study is prompt engineering and Multi-Agent Systems (MAS). This approach involves crafting specific prompts that elicit tailored responses, enhancing the overall learning experience. To maintain student engagement and motivation, the system incorporates educational techniques such as gamification, interactive simulations, and adaptive feedback mechanisms. Furthermore, the ITS is designed to recognize cues for fatigue and distraction by analyzing patterns in student interactions, such as response times and engagement levels. The effectiveness of the system is validated through extensive testing with AI-simulated students of varying proficiency levels, providing valuable data for refining prompts and improving personalization. Overall, the paper demonstrates how integrating GenAI technology can create more efficient and flexible educational environments, addressing the diverse needs of learners and redefining the landscape of personalized education.