COMPEL Glossary / retrieval-augmented-generation-rag
Retrieval-Augmented Generation (RAG)
Retrieval-Augmented Generation is a technique that enhances AI model responses by first retrieving relevant information from external knowledge sources -- databases, document repositories, knowledge bases -- and then using that information as context for generating more accurate, grounded answers.
What this means in practice
RAG addresses the hallucination problem by giving the model access to verified factual information rather than relying solely on patterns learned during training. For enterprises, RAG enables AI assistants that can answer questions using the organization's own documents and data while reducing the risk of fabricated responses. RAG architectures require careful governance of the knowledge sources being retrieved, access controls to prevent unauthorized information disclosure, and monitoring to ensure retrieval quality remains high over time.
Related Terms
Other glossary terms mentioned in this entry's definition and context.