Part ofAI in product and site
RAG assistants
RAG stands for retrieval augmented generation and means the model retrieves relevant passages from your own documents before composing an answer, rather than answering from memory. It is the single most important technique for making an assistant usable in production, because every claim can then be traced to a document that actually exists.
Retrieval decides the quality
Almost all poor RAG answers come from the wrong passages being retrieved, not from a weak model. The work therefore sits in the retrieval step: how content is chunked, how results are ranked and how many passages are passed along. Changing model is the last adjustment, not the first.
Being able to say it does not know
An assistant has to be able to decline. When nothing sufficiently relevant is retrieved, the answer should be that the information is not available, with a route through to a person. Without that rule the model fills the gap with something plausible, and that is where trust is lost.
Permissions travel with retrieval
In an internal tool the assistant must not surface a document the user is not allowed to read. Permission checks therefore happen during retrieval, so passages are filtered out before the model sees them. Filtering the answer afterwards is not a control, it is a hope.
Common questions about this service
Can the assistant still make things up?
The risk drops sharply but does not vanish. With retrieval from your documents, a citation in the answer and a rule to decline when the evidence is missing, the remaining errors usually sit in the phrasing rather than the facts. That is why the prompt, the retrieved passages and the answer are all logged, so deviations are found in review rather than by a customer.
Do we have to send our documents to OpenAI or Anthropic?
The retrieved passages go along with the call, yes. Where that is unacceptable there are two routes: vendor agreements where data is not used for training and is processed inside the EU, or a model running in your own environment. The second costs more to operate and gives weaker answers, and that trade-off should be made deliberately rather than assumed.
More in the same area
AI search and vector search
Search that understands what the visitor meant, not just which words they typed.
RAG assistants
Assistants that answer only from your documents, with the source cited.
Product content at scale
Product copy and metadata generated from your structured data, not from nothing.
