Title:	AI-POWERED PEDAGOGY: GENERATING RELIABLE AND CONTEXT-AWARE EDUCATIONAL CONTENT FROM INSTRUCTOR-CURATED MATERIALS
Author(s):	Mikhail George Samir Youssef Gebrail and Ahmed M. H. Abdelfattah
ISBN:	978-989-8704-71-9
Editors:	Paula Miranda and Pedro Isaías
Year:	2025
Edition:	Single
Keywords:	AI in Education, Instructor-Controlled AI, Source Attribution, Multimodal Personalized Learning, Trustworthy AI
Type:	Full Paper
First Page:	91
Last Page:	98
Language:	English
Cover:
Full Contents:	if you are a member please login
Paper Abstract:	The use of Large Language Models (LLMs) in education presents a significant challenge for educators, as the reliability of AI-generated content is often compromised by hallucinations and unverifiable sources. This diminishes instructor control over learning materials and creates a risk of students building knowledge on flawed foundations. This paper introduces "Studymate", a proof-of-concept implementation of a model-agnostic framework designed to give control back to the instructor by providing trustworthy, contextually precise academic support that operates exclusively on their provided course materials. The implemented system features a multi-step pipeline to deliver varied educational outputs, including a conversational Question/Answer (Q/A), multi-modal PDF summaries with generated images, and podcasts. All information presented is grounded in the course-specific knowledge base, with precise citations to the source document and page number, ensuring verifiability. The main goal achieved by the system is that it restores control to educators and provides students with a dependable and effective learning aid that is reliable, correct, and directly controlled by their instructors. This work contributes a practical, model-agnostic framework for creating trustworthy AI-assisted educational tools and addresses the critical need for source attribution in AI-generated academic content.
	Go Back