Microsoft Builds Fabric Into a Business Context Layer for Copilot and AI Agents

Microsoft has announced a set of Microsoft Fabric and Azure database updates aimed at helping AI tools work with trusted business data and definitions. Revealed at the European Microsoft Fabric + SQL Community Conference in Barcelona, the changes span Copilot, governed data sharing, application development, data engineering and database management—with several features generally available, in […]

Microsoft Fabric connects AI agents and Copilot.

In This Article

Share this

Microsoft has announced a set of Microsoft Fabric and Azure database updates aimed at helping AI tools work with trusted business data and definitions. Revealed at the European Microsoft Fabric + SQL Community Conference in Barcelona, the changes span Copilot, governed data sharing, application development, data engineering and database management—with several features generally available, in preview or planned for preview.

Key takeaways

Microsoft’s announcements extend Fabric’s role beyond analytics, connecting business context with AI experiences and operational tools. For enterprise teams, the central opportunity is to make data more usable while retaining governance and control.

  • Fabric IQ in Copilot Chat and Cowork is generally available, grounding business answers in governed Power BI data.
  • IQ Sharing and Database Hub are in preview; new agentic data engineering features are designed to keep engineers in control.
  • Fabric Apps is adding capabilities to build applications on governed data, while SQL Server on Azure Local is generally available.

Fabric IQ brings Power BI context to Copilot

Fabric IQ connects Copilot Chat and Cowork to Power BI semantic models, including trusted metrics, relationships and business definitions. Users can ask business questions and receive answers grounded in governed data; Cowork can use those insights in multistep work. Microsoft says the capability is generally available and preserves existing access controls. Integration with Copilot’s Code experience is planned through the Frontier program.

For organizations, grounding AI in established definitions can reduce confusion over which figures to use and help teams move from reconciling reports to acting on results. That depends on having well-maintained semantic models and governance—a core consideration for enterprises investing in Fabric and Power BI. Spargent’s focus on unifying data and improving reporting workflows reflects the implementation work behind making those foundations useful at scale.

Sharing and building on governed data

IQ Sharing, now in preview, is intended to let organizations share Fabric data and business context with other teams and external partners while maintaining governance over underlying assets. Microsoft says it will begin with Fabric IQ artifacts and context, with plans to expand. The feature complements OneLake mirroring and shortcuts for sharing data across environments.

Fabric Apps is also adding connectivity, transactional workloads, backend logic, storage and policy-controlled access. The capabilities are intended to help developers build applications directly on governed Fabric data, including warehouses, SQL databases, lakehouses and Power BI semantic models, rather than duplicating data across systems.

Agents and database operations

New agentic data engineering features let engineers set an intended outcome and boundaries for an agent to plan and execute work, validate results and refine its approach. The model gives teams a way to automate complex engineering tasks while retaining human-defined limits and oversight.

Database Hub in Fabric is in public preview as a central control plane for SQL Server, Azure SQL, PostgreSQL, Cosmos DB and SQL databases in Fabric. Specialist SQL and PostgreSQL agents for Database Hub and Visual Studio Code are expected in public preview, aimed at troubleshooting, performance analysis and optimization. Separately, SQL Server on Azure Local is generally available, with disconnected operations in preview for environments with limited or no cloud connectivity.

What enterprises should watch

Microsoft’s strategy is to make business context available across data, applications and AI assistants, rather than treating models as standalone tools. The practical value will depend on data quality, clear ownership of business definitions, access policies and interoperability with systems beyond Microsoft. Leaders evaluating the updates should assess how they fit existing architecture and whether governed data foundations are ready to support reliable AI workflows.

Sources

Spargent Analytics Logo Microsoft Fabric Consulting services

Spargent Analytics

Microsoft Fabric consulting, implementation, analytics modernization, and long-term support for enterprise data teams.

Microsoft Fabric
Project Review

Free Expert Session
Need help turning this insight into a Microsoft Fabric roadmap?

Spargent Analytics can help you design, implement, migrate, and optimize Microsoft Fabric solutions that bring your data, analytics, AI, and business intelligence into one secure and scalable platform.

More insights

Continue with related Microsoft Fabric articles.

Fabric Eventstream Latency: How to Trace the Delay

Fabric Eventstream latency can begin before the stream receives an event, during transformation, at the destination, or in reporting. Trace

Microsoft Fabric Semantic Models for Finance Teams

Finance teams quickly lose trust when revenue figures vary between reports. In a Microsoft Fabric semantic model, trust depends on

Unsealed filings expose AI scraping’s legal and business risks

Unsealed filings from The New York Times’ copyright case against OpenAI and Microsoft reveal that senior executives privately characterized AI

Start a Conversation

We will get back to you within 24 hours with proposal to set up intro call.