Author: Miquella de Boer - Principal Product Manager Lead
Organizations do not have a data problem. They have a fragmentation problem.
The data that matters most to AI is often already present across the enterprise, but discovering the right asset, understanding whether it is trusted, and connecting it into an agent workflow still takes too much effort. Users should not have to jump between products, hunt for IDs, or recreate context that already exists elsewhere. Which is why we are bringing the OneLake catalog natively into Foundry.
With this experience, users can browse and select OneLake data directly from within Foundry through a native catalog experience. Instead of manually wiring up data sources one at a time, they can discover relevant assets in place, understand what they are looking at, and turn them into knowledge sources without leaving the flow they are already in.
This is a meaningful step forward in how data and AI come together across Microsoft. OneLake already serves as the unified data foundation across Fabric and beyond. By bringing the catalog directly into Foundry, we are making that foundation available where AI builders are already working. The result is a more natural path from governed enterprise data to grounded AI experiences.
Figure: Browse OneLake in Foundry.
Why this matters
The path from enterprise data to grounded AI can still feel too fragmented. The current integrated catalog direction is designed to eliminate manual URL retrieval and make governed data selection feel native inside Foundry.
Instead of starting with technical setup, users start with discovery. They can browse the data available to them, understand what they are selecting, and move directly into creating a knowledge source.
From governed discovery to grounded AI
Users can browse relevant assets in context. They can evaluate signals like ownership, endorsement, sensitivity, and location before deciding what to use. And once they find the right source, they can move straight into the next step in Foundry instead of switching tools or reconstructing context somewhere else.
This integration is about more than convenience. It reflects a broader direction toward a more unified experience across data and AI. That means less fragmentation, less duplicated effort, and a clearer path from data discovery to AI grounding.
As we continue to bring Fabric and Foundry closer together, this is a foundational step toward a simpler model where data, knowledge, and AI assets are easier to discover, trust, and reuse across Microsoft experiences.
How to get started
Prerequisites
- A lakehouse in Fabric - If you don’t have a lakehouse, follow the steps in Create a lakehouse with OneLake.
- A Foundry project - If you don’t have one, follow the steps in Create a project.
- An Azure AI Search service at the Basic tier or higher - If you don’t have one, follow the steps in Create an Azure AI Search service.
- The search service must be in the same tenant as your Fabric workspace.
- Create and assign a managed identity for the search service. To create a managed identity, you must be an Owner or User Access Administrator roles. To assign roles, you must be an Owner, User Access Administrator, Role-based Access Control Administrator, or a member of a custom role with Microsoft.Authorization/roleAssignments/write permissions.
Create a OneLake connection in Foundry
- Sign in to Microsoft Foundry. Make sure the New Foundry toggle is On through Microsoft Foundry (new).
- Open the preferred project.
- Select Build from the navigation menu, then select Knowledge from the left pane.
Create knowledge base in Foundry.
- Select your AI Search resource.
Select Azure Search resource.
- Select Create a knowledge base.
- Select Microsoft OneLake as the knowledge type, then Connect.
- Browse the OneLake catalog and select the item you want to add.
- Select Create.
- Select Save knowledge base.
Ready to try it out?
Native OneLake catalog integration in Foundry removes the friction between governed data discovery and grounded AI. Your agents can stay anchored on enterprise data with the same security and governance controls you already rely on in OneLake and Fabric.
Learn more, with step-by-step instructions in the OneLake for Microsoft Foundry documentation.