Of all the technical concepts educators encounter when they start working with AI, Retrieval-Augmented Generation (RAG) is the most important, not because it is the most complex, but because it explains the most. It explains why AI sometimes makes things up. It explains why AI produces generic, unhelpful responses when given no context. And it explains how teachers can transform AI from a generic content generator into something that actually sounds like them and trust AI as a collaborative partner is guiding students towards learning outcomes. 

This session translates RAG from a technical architecture into a practical teaching framework. Attendees will leave understanding why AI hallucinates (and that it is not random), the three concrete levers they control to reduce it, and how teachers who clarify their own instructional standards are already doing the hardest part of building an AI knowledge base. No coding required.