Microsoft Fabric and Azure AI Foundry announced a key enhancement enabling enterprises to use files stored in OneLake as a native knowledge source for AI agents – allowing unstructured and semi-structured data to power AI-driven business processes without duplication or complex data pipelines.
With this integration, companies can connect Foundry directly to their OneLake lakehouse, index files such as documents, images, logs or CSV/JSON files via the built-in indexer, and use the indexed content inside AI agents for retrieval-augmented generation (RAG), search, summarization or Q&A. This approach ensures that agents operate on the same governed, security-compliant data store used for analytics and reporting, rather than requiring separate AI-specific data silos.
Microsoft states that this capability dramatically simplifies enterprise adoption of AI-eliminating redundant data transfers and enabling secure, governed use of enterprise content in AI workflows.
Also Read: Supabase and AWS Partners to Accelerates Developer Scale
This update comes as part of Microsoft’s broader push to unify data, analytics, and AI – offering organizations the tools to transform raw enterprise data into actionable intelligence, supported by a scalable, governed foundation.