RAG

RAG is an AI term used to describe systems that combine a language model with external sources of information. Instead of relying solely on what the AI was trained on, a RAG system first retrieves relevant documents, webpages, or database entries, then uses that information to generate a more accurate response.

AI researchers introduced the concept in the late 2010s, and the term became popular in the mid-2020s as businesses began building smarter chatbots, AI assistants, and search tools. Today, developers, AI engineers, tech companies, and startup communities working with large language models and enterprise AI tools widely use the "RAG" acronym. They often mention it in meetings, online discussions, and developer chats when discussing ways to make AI more current, reliable, or connected to company-specific knowledge.