RAG (Retrieval-Augmented Generation)
RAG is a technique where a language model answers based on the business's own documents — instead of only its general training.
Also known as: retrieval augmented generation, context-based AI
RAG (Retrieval-Augmented Generation) is a technique where a language model retrieves relevant excerpts from a knowledge base or document archive before generating its answer. Instead of relying on the model's general training, the answer is grounded in actual documents — with the option to cite the source. RAG solves two problems at once: the AI gives answers based on your business's data, and you can see where each answer came from. Common use cases are internal knowledge search, customer service chats grounded in product documentation, and contract search.
In Norwegian context
For Norwegian businesses, RAG is often the first choice when building an AI assistant that "knows" the company's own documents — because data stays on Norwegian infrastructure and access control can be enforced per document.
Read more in the in-depth article on this topic.
Related terms
- AI integration — AI integration is the process of wiring language models, RAG, or predictive models directly into a business's existing systems and workflows.
- LLM (Large Language Model) — An LLM is a large language model trained on enormous text volumes that can generate, summarise, and analyse text in a human-like way.
- Embedding — An embedding is a numerical representation of text (or image) that lets AI systems compare semantic similarity.
- Vector database — A vector database is a database optimised for storing and searching embeddings — the foundation of RAG systems and AI search.