If you haven’t already, check out Arun Ulag’s hero blog “Microsoft Build 2026: Building Agentic Apps with Microsoft Fabric and Microsoft Databases” for a complete look at all of our Microsoft Build announcements across our Fabric and database offerings.
_______________________________________
Microsoft Build 2026 brings the next generation of intelligent analytics to Power BI—where agents do the building, apps deliver insights everywhere, and the platform gets more powerful for every user.
Analytics teams have moved past asking whether AI belongs in their BI stack. The priority now is making that BI stack ready for AI. At Build 2026, we answer that need with four announcements that position Power BI as a platform for everyone: analysts, developers, business users, and the agents that support them.
Leverage our agent skills for Power BI for end-to-end agentic analytics
AI without semantic context is a confident guesser. Power BI provides business logic that ensures trustworthy analytics. Put a language model in front of raw tables and it will produce answers that look right—until they’re not. The difference is a well-curated semantic model: measures, relationships, hierarchies, and definitions that reflect how your organization works.
Agent Skills for Power BI (Preview), announced at Build, provide true end-to-end agentic development to Power BI, from raw data to semantic models to interactive reports. Describe to your AI agent what you need, even from a screenshot, and it can create models, generate reports, and iterate on visuals, all in one agentic workflow.
Figure: Agent Skills to create a semantic model and a report from scratch.
Build enterprise web apps on semantic models
Announced at Build, Fabric Apps introduce a new AI-first way to build custom web apps with Microsoft Fabric as the backend. For analytics teams, this means developers and their AI coding agents now have an accelerated path to building enterprise-grade data apps directly on their semantic models. Organizations can create and deploy operational data apps that leverage the same trusted business logic and the same governance as the rest of their analytics stack.
From financial planning to inventory management to pricing optimization, or really any app you can describe, your coding agent can build a custom-tailored app based on your specifications and semantic model in just a few prompts. And it's not just about polished UI. Coding agents can trivialize features that are used to take real engineering effort, like persona-specific views, custom calendar interfaces, bespoke business logic, and more.
Figure: Fabric data app on top of semantic model with Outlook calendar integration.
Check out our blog post Building in the Agentic Era with Power BI and Fabric about the latest announcements in agentic development in Power BI and Microsoft Fabric.
Delivering curated and customized analytic experiences at scale with Org apps
Organizational apps have always been one of the most effective ways to distribute analytics at scale. In the coming weeks, we’ll announce the general availability of org apps, bringing a polished, stable, enterprise-ready distribution experience to our customers.
With org apps re-envisioned for Fabric, teams can create multiple org apps per workspace and distribute unique content experiences to large audiences across their organization. Each org app serves as a curated collection of Power BI and Fabric items, ranging from reports to notebooks to real-time dashboards and more, all within an easy to use, navigable, branded, and secure experience.
With this milestone, we are introducing audience control, a feature frequently requested during preview. Audiences in org apps allow you to define which items are visible to specific groups of users within the app.
Org apps let you securely distribute analytics across your organization at scale within customized experiences.
Figure: Org app with audiences.
Making custom logic reusable, consistent, and easy to scale with User-Defined Functions
Power BI's semantic layer is where AI interprets business logic, and that layer is becoming more expressive.
With DAX user-defined functions, teams can define reusable, parameterized logic once and apply it consistently across measures, calculated columns, visual calculations, and even other functions. That means less duplicated code, easier maintenance when business rules change, and cleaner models that are simpler for both humans and AI systems to understand.
In practice, user-defined functions turn the semantic model into a more durable foundation for analytics—one where complex logic can be standardized, shared, and scaled without losing clarity. That matters not just for model authors, but for the broader direction of Power BI: as more experiences become agent-driven, reusable semantic logic becomes the bridge between trusted business definitions and the applications, reports, and workflows built on top of them.
The road ahead
These announcements share a common thread: Power BI is becoming the semantic foundation that makes AI trustworthy and the delivery platform that gets insights to every user. Whether you're building a governed semantic model, deploying an app to thousands of business users, or letting an agent write your next data app, the infrastructure is here.
Catch these announcements in action in our on-demand session ‘Build Agentic Analytics with Power BI and Microsoft Fabric’, available in the Build catalog.
- Learn more about all the Fabric announcements in Microsoft Build 2026: Building Agentic Apps with Microsoft Databases and Microsoft Fabric.
- Explore the Power BI documentation to dive deeper into these capabilities.
- Join our community to share your feedback and meet others on the same journey.