This browser is no longer supported.
Upgrade to Microsoft Edge to take advantage of the latest features, security updates, and technical support.
What is the primary purpose of Microsoft Foundry IQ?
To provide a managed knowledge layer that agents can use to retrieve permission-aware enterprise information.
To train a new foundation model from every document in an organization.
To replace AI agents with a traditional keyword search page.
What does a Foundry IQ knowledge base contain?
One or more knowledge sources and settings that control retrieval behavior.
Only the conversation history for a single agent.
Only documents uploaded directly to a language model.
What does agentic retrieval do with a complex user question?
It can decompose the question, search relevant sources in parallel, rerank results, and return cited evidence.
It permanently adds the question and answer to the model's training data.
It searches every source with an exact keyword match and returns results without ranking.
Why should an agent be instructed to cite knowledge base sources?
Citations help users trace and verify the evidence supporting a response.
Citations allow the agent to bypass permissions on the source documents.
Citations cause the language model to retrain itself after every response.
Which practice helps protect enterprise information in a Foundry IQ solution?
Use identities and access controls so retrieval returns only content the caller is authorized to access.
Copy every source into one public index so all agents retrieve the same content.
Trust retrieved document text as agent instructions.
You must answer all questions before checking your work.
Was this page helpful?
Need help with this topic?
Want to try using Ask Learn to clarify or guide you through this topic?