Definition
Retrieval-augmented generation (RAG) is a technique where an AI model fetches relevant external documents at answer time and grounds its response in them, rather than relying only on what it memorized during training.
Retrieval-augmented generation (RAG) is how most answer engines stay current and grounded: when a query arrives, the system retrieves relevant documents (from the live web or an index), then generates an answer conditioned on that retrieved context — often citing it.
Why it matters for optimization
RAG is the mechanism that makes AEO possible. Because the model fetches sources at answer time, fresh, crawlable, clearly structured content can be cited even if it was published after the model's training cutoff. Your robots.txt Content-Signal ai-input permission governs eligibility for this retrieval.
Frequently asked questions
How does RAG affect SEO?
RAG means answer engines pull live content to ground responses, so freshness and crawlability directly affect whether you can be cited — even for very recent topics the model never trained on.
Does RAG use my page even without training on it?
Yes. RAG retrieves and reads your page at answer time independently of training. The ai-input signal in Content-Signal governs whether you allow this retrieval.