# What is RAG?

Official page: https://aitje.com/en/knowledge/wat-is-rag

> RAG is a method whereby a language model first retrieves relevant information from documents or a knowledge base before providing an answer. As a result, answers are often more specific, better substantiated and more useful for your own organisation.

### What does RAG stand for?

RAG stands for Retrieval-Augmented Generation. First, the system retrieves relevant information from a knowledge base, document collection or other source. Only then does the model use the context it has found to construct an answer.

### What happens from a technical perspective?

In a RAG system, a query is first translated into a search. Relevant text passages are then selected and sent to the model. The model therefore responds not only on the basis of its training, but also on the basis of sources retrieved at that moment.

### Why is this valuable?

A standard LLM does not automatically know what is contained in your manuals, policies or project documents. With RAG, the system can provide answers based on your own information. This makes AI much more useful for internal knowledge enquiries, document support and customer-focused processes.

### What determines quality?

The quality depends on the source data, document structure, search logic and whether the correct passages are selected. Poorly structured documents or out-of-date information will also result in substandard answers in a RAG setup. RAG is therefore not a panacea, but a process that must be properly organised.

### When is RAG not enough?

RAG helps with knowledge retrieval, but does not solve everything. For complex workflows, decision logic, rights management or actions in other systems, you often need additional layers, such as validation, human verification or agent logic. RAG is often a strong foundation, not the complete product.

## Contact

[Contact AITJE](https://aitje.com/en/contact) — contact@aitje.com