2026 ASEE Annual Conference & Exposition

The use of case studies is an established learning intervention for teaching students about the complex and contextual nature of how technology impacts society, and vice-versa. In this paper we present the design and initial implementation of GENCAS (GenAI-Based Case Study Assessment and Feedback System) an interactive, AI-enabled educational platform designed to evaluate and enhance student understanding of real-world case studies. Leveraging our prior work on using the Boeing 737 Max incident as a case study, we designed an AI-based instantiation where we used transcripts from prior group discussions to train the AI system to assess and provide feedback to students. Our system combines a lightweight user interface with a Retrieval-Augmented Generation (RAG) workflow to deliver context-aware, personalized feedback on student responses. A student goes through a structured question flow and submits open ended answers. For each answer, the application retrieves the most relevant prior discussion fragments and constructs a composite prompt consisting of the current question, the student’s answer, and contextual snippet. This prompt is sent to an OpenAI model (gpt‑4o‑mini) to generate formative feedback that references authentic peer reasoning rather than generic rubric language. GENCAS demonstrates a practical, privacy-conscious pattern for augmenting assessment with retrieval-grounded generative AI to produce timely, context-rich feedback that scales beyond traditional manual evaluation. We present findings from user evaluation where most users rate the system positively.