RAG (Retrieval-Augmented Generation)
A technique where an AI looks up real documents or sources before answering, instead of relying only on what it memorized during training.
01 Definition
RAG combines an AI model with a search step: instead of answering purely from memory, the system first retrieves relevant, up-to-date documents or web pages, then generates an answer grounded in that retrieved content — often with citations.
02 Why It Matters
RAG is why some AI tools can cite sources and stay current on recent events, while a plain chatbot can only draw on what it learned during training (which has a cutoff date).
03 Example
Perplexity uses RAG to answer questions with live, cited web sources instead of relying only on trained-in knowledge.