# Knowledge Base

> A knowledge base is an organized collection of information a business uses to answer questions and do its work, such as help articles, SOPs, policies, product documentation and resolved support tickets. In AI systems, the knowledge base is the content an assistant retrieves from to ground its answers, so its accuracy, structure and freshness directly limit the quality of the AI's responses.

## Why it matters for your business

Many AI assistant projects turn out to be knowledge base projects in disguise. If policies are outdated, scattered across drives or contradict each other, the AI will repeat those problems quickly and at scale. Before building, it pays to identify the authoritative source for each topic, retire duplicates, assign owners and set a review schedule. A maintained knowledge base also helps human staff and new hires, so the cleanup has value even before any AI is added.

## Example

An IT services company wants an assistant that answers technicians' questions. The team finds three versions of the onboarding checklist and outdated firewall procedures in a shared drive. They consolidate the content in one wiki, assign an owner to each page and archive the rest. The assistant only indexes pages marked current, and each answer links the page it used.

Related terms: https://www.lumeor.studio/glossary/rag-retrieval-augmented-generation/, https://www.lumeor.studio/glossary/grounding/, https://www.lumeor.studio/glossary/chunking/, https://www.lumeor.studio/glossary/chatbot/, https://www.lumeor.studio/glossary/embeddings/

---

Source: https://www.lumeor.studio/glossary/knowledge-base/