An Explainable AI Assistant for Introductory Programming Education: Improving Feedback Reliability with Instructor-AI Collaboration

This paper presents an AI-driven classroom assistant designed to enhance feedback reliability in introductory programming courses by utilizing an explainable AI model. The assistant analyzes student code, identifies logical errors, and provides feedback based on instructor-defined misconceptions, thus ensuring alignment with pedagogical knowledge. The framework was evaluated through expert assessments and classroom deployment, demonstrating its effectiveness in delivering accurate, instructor-verified feedback and improving student perceptions of usability, which is crucial for practitioners aiming to integrate reliable AI tools in educational settings.