As organizations scale their data and AI investments, many are adopting a multi-platform approach so teams can use the tools that best fit each project. But this flexibility comes at a cost: duplicated data, fragmented governance, and complex data movement across systems.
The next phase of data platforms isn’t just about connecting systems; it’s about sharing a common data foundation and, increasingly, a context foundation. Instead of aligning on a single engine, organizations can align on a single copy of data while enabling teams to use the tools they use.
With today’s updates, Microsoft Fabric and Azure Databricks take a major step forward, enabling you to use Microsoft OneLake as a shared data layer that both platforms can read and store in natively. This means Azure Databricks customers can use OneLake as their native storage option, ensuring data lives in one place with a single copy, but can be accessed in either platform and engine.
Expanding Azure Databricks interoperability with bi-directional OneLake access, now available
With these updates, Azure Databricks and Microsoft Fabric enable true bi-directional interoperability through OneLake. Customers can now read and store the same data from either platform without duplication or complex pipelines.
- Now in beta: Azure Databricks can store Unity Catalog managed tables natively in OneLake
- Now generally available: Azure Databricks can read data stored in OneLake
With support for reading and storing directly to OneLake, Azure Databricks customers can now use OneLake as a native storage layer for their Delta tables without managing separate storage systems. This lets you store data in OneLake while using either Fabric or Azure Databricks for each project.
Because there is a single copy of data, changes made in one environment are immediately reflected in the other. This eliminates the need for data movement, reduces duplication, and simplifies how organizations manage and govern their data estate.
Learn more about these integration’s in Premal Shah’s “Extending interoperability: Azure Databricks stores directly in OneLake” blog post.
Building an effective AI strategy with Azure Databricks and Microsoft OneLake
For Azure Databricks customers, this new interoperability means using OneLake as the native store for your data. So why does this matter?
As organizations move from traditional applications to AI‑powered, multi‑agent systems, the advantage is shifting beyond the specific model you deploy. It now lies in the intelligence and context that allow agents to understand how your business is run, the state of your business, and your institutional knowledge to help take meaningful action.
That’s the challenge OneLake is designed to address. OneLake is designed as the single, multi-cloud data lake to fuel all your AI projects. Microsoft OneLake can supports your AI teams through each of the four critical steps of preparing your data for AI:
- your data estate
- Processing data
- Curating semantic meaning
- Empowering AI agents to act
Unify your data estate with shortcuts and mirroring in OneLake
With OneLake, you can access your entire multi-cloud data estate from a single data lake that spans the entire organization. It connects data across clouds and on‑premises systems easily using zero‑copy, zero‑ETL shortcuts and mirroring, so teams work from a single, governed copy of data. Whether your data is in Azure, Amazon Web Services, Google Cloud, on-premises SQL Server, or platforms like SAP, Dataverse, Snowflake, or Databricks, you can quickly connect it to without moving or duplicating it.
Once your data is connected to OneLake, it becomes easily discoverable through the OneLake catalog, where data assets are listed along with their governance and security metadata. The OneLake catalog is integrated into the tools your teams already use—such as Power BI, Microsoft Teams, Microsoft Excel, Microsoft Copilot Studio, and Microsoft Foundry—so users can find and use data in the tools they already use.
Process data with full choice of any data platform
AI agents are only as reliable as the data you feed them. Before enterprise data can train or ground an AI agent, the data often needs to be cleaned, curated, and validated so the agent is working from consistent, trusted information.
With native support for Delta Lake and Iceberg data formats, Microsoft OneLake works seamlessly with any analytics engine in Microsoft Fabric, Azure Databricks, or even Snowflake. And as your data moves through pipelines, notebooks, warehouses, or streaming jobs in any platform, it stays in OneLake as a single copy—governed, discoverable, and continuously available to power downstream analytics, reporting, and AI applications.
Curating semantic meaning with Microsoft IQ
Once data is prepared, the next challenge is making it understandable. Many organizations lack a shared layer of business context, forcing every agent to relearn how the business works from fragmented data. Fabric IQ addresses this gap.
With data unified in OneLake, Power BI’s industry-leading semantic models then provide structured representations of your data for trusted business intelligence, serving as an ideal foundation for training agents. Ontologies in Fabric IQ extend semantic models by adding operational context. They define business entities, relationships, properties, rules, and actions, and connect to live signals from Fabric Real-Time Intelligence. This can help you power a continuous operational loop where people and agents observe live signals, reason over shared context, and take governed action in the moment.
But Fabric IQ is only one slice of the context you can use to fuel your agents. Fabric IQ is one pillar of Microsoft IQ that also brings together three other interconnected capabilities: Work IQ captures how work happens; Foundry IQ enables agents to discover and reuse knowledge; and Web IQ adds real-time global context from the web.
Microsoft OneLake is natively integrated with the entire Microsoft IQ stack, enabling you to unify enterprise intelligence into a shared foundation built to activate AI agents.
Empowering AI agents to act
Finally, data in OneLake can power agents in the AI platform you choose. The OneLake catalog natively appears in Microsoft Foundry for example, giving you the ability to discover trusted data, explore rich metadata, and connect it directly to AI solutions. The catalog is embedded within Foundry’s Knowledge experience, making it simple to move from data discovery to AI development in a single workflow.
Securing and governing your data and agents
OneLake provides built-in governance and security so your data remains discoverable, controlled, and compliant.
Learn more
With OneLake as a shared data foundation, you don’t have to choose between platforms—you can bring them together. By keeping a single copy of data that works across Azure Databricks and Microsoft Fabric, you can simplify your architecture, reduce duplication, and focus on building AI solutions that scale.
Explore how Microsoft OneLake can unify your data estate and support your AI strategy across platforms.
Microsoft Employee