Content revision 4e771eaf0390cc87b6c7c6fd8469f3508f6ceeec7119472e6c1ccf17e2e1e5d2

## Where does the assistant need more knowledge?

AI for IT Operations
Source ID ai-it-operations:where-does-the-assistant-need-more-knowledge
Original https://www.henryw.me/#/ai-it-operations#where-does-the-assistant-need-more-knowledge

Before collecting more documents, I needed to identify what information was missing. Having material on a topic does not tell us which questions it can answer. Documentation on certificates might explain how to renew one while leaving out how to get an application to use its replacement. Checking those two questions separately makes the missing information explicit and gives the next search something specific to find. Questions the assistant had been unable to answer provided one way to discover gaps. I also used AI to propose useful questions beyond those people had already asked, so the collection could develop ahead of demand. A proposed question still needed to be checked against the existing material. A new question does not necessarily require new knowledge. For example, documentation collected on hosting a Python web app may already explain how to host a REST API.

Whether a question has been asked and whether the collection can answer it are therefore separate considerations. The matrix below crosses them to show four cases. The two missing cases supply questions for further research. Try predicting a question, then follow the search for the information it needs.