RAG (Retrieval-Augmented Generation)
An AI architecture in which a model retrieves relevant information from a knowledge base or the live web before generating its response.
Explanation
RAG describes an AI setup where a model pulls in outside information before writing an answer, rather than relying only on what it learned during training. Facebook AI introduced the term in 2020-era NLP research, and it moved into everyday marketing vocabulary through 2025 as GEO, AEO, and LLMO practitioners began explaining it to clients as the mechanism behind AI search visibility. Google AI Overviews and Perplexity both use RAG-style retrieval before generating their summarized answers. Marketers now build content that is fresh, well-structured, and clearly sourced specifically so it can be retrieved by these systems.
Usage context / examples
Google AI Overviews and Perplexity both rely on RAG-style retrieval before generating their summarized answers.
Related terms
LLM, AI Orchestrator, GEO, AEO, LLMO
Turkish definition
RAG, bir yapay zeka modelinin yanıt üretmeden önce bilgi tabanından veya canlı web'den ilgili bilgiyi getirdiği bir mimaridir.
Source: MMA Academy Sözlük