Knowledge Base
Also called KB
The curated pile of documents a system searches to answer questions with real facts.
Think of it like
The company handbook everyone points to instead of guessing the policy.
Example
A support assistant is wired to a knowledge base of help articles, so its answers cite your actual docs, not the open internet.
How it actually works
A knowledge base is the source of truth a retrieval system draws on — help docs, policies, product specs, past tickets. Its quality caps the whole system: stale, contradictory, or missing content produces confident-but-wrong answers no matter how good the model is. Curating and refreshing it is ongoing work, not a one-time load.
For product teams
The answer quality ceiling; investing in clean, current content beats most model tweaks.
For engineers
The corpus feeding ingestion and retrieval; freshness and dedup are operational, not one-shot, concerns.
Related
- RAG — What a RAG system retrieves from.
- Document Store — Stored and chunked for retrieval.
Read anything AI without the jargon
Look up any term in plain English, or save terms as you read with the free Chrome extension.
Open DecoderAdd to Chrome