At Microsoft Ignite 2026, we are announcing new integrations that bring Azure Databricks intelligence into Microsoft 365 Copilot and Copilot Cowork, connect business meaning through Genie Ontology alongside Microsoft IQ, and bring Genie agents into enterprise agent oversight. The goal is to help customers put enterprise data to work through AI, with the business context and governance that data requires.
For business users, that means asking questions in familiar Microsoft tools. For data teams, it means extending the reach of the data and business definitions they manage. For agent builders and IT teams, it means connecting those experiences to enterprise intelligence and oversight.
These updates build on Azure Databricks, the jointly engineered Microsoft and Databricks service that brings data engineering, analytics, and AI together. They extend that foundation across a Microsoft ecosystem that includes Foundry, Fabric, and Copilot.
Bring enterprise data into the task at hand
A business question rarely ends with a query result. A sales leader investigating a change in performance needs an answer they can use in a review, a discussion, or a decision.
The Genie One plugin, built on Model Context Protocol (MCP), lets Microsoft 365 Copilot and Copilot Cowork call Genie for Databricks data questions as part of a broader task. Copilot plans the work, selects tools, and composes the final response. Genie contributes the intelligence needed for the data question.
Copilot sends Genie a scoped question, not the entire user prompt. The user can keep working in Copilot while Genie handles the relevant part of the request.
For people who want to explore their data directly, the Genie app in Teams and Microsoft 365 Copilot provides a conversational experience with Genie. Whether the starting point is a direct data question or a larger Copilot task, Azure Databricks intelligence becomes accessible within the tools people already use.
Put business meaning to work beyond the Lakehouse
Data teams invest in more than pipelines and tables. They establish what a metric means, which sources are authoritative, and who can access the underlying information. Those decisions need to remain relevant when AI becomes another consumer of enterprise data.
Genie Ontology contributes the business meaning of Azure Databricks data, including metrics, semantics, and usage, with access gated by Unity Catalog permissions.
That intelligence complements Microsoft IQ. Work IQ, Fabric IQ, Foundry IQ, and Web IQ provide shared context about how people work, how the business runs, what the organization knows, and what is happening on the web. Genie Ontology adds the meaning of the organization's Databricks data.
For a question about revenue, the intended metric matters as much as the result. Connecting work context with business meaning helps bring the data team's definitions into the AI experience, while Unity Catalog permissions continue to gate access to Azure Databricks data.
The benefit is a wider role for the governed data foundation: the work data teams do can support conversations and tasks beyond a traditional analytics interface.
Connect agents to governed Databricks intelligence
An agent that coordinates work needs a way to ask the right system the right question. The Genie One MCP integration gives Copilot a defined path to Genie for Databricks data questions, rather than treating every request as a generic prompt.
For teams building custom agents, these updates build on the previously announced workspace-wide MCP endpoint for Copilot Studio. That endpoint lets an agent work across connected Genie spaces and Unity Catalog datasets through one workspace connection, preserving the curated context in Genie spaces and enforcing data permissions at query time.
The distinction matters: the Genie One plugin brings Databricks intelligence into Copilot-coordinated work, while the Copilot Studio integration supports teams authoring their own agents. Both extend the usefulness of the governed Azure Databricks foundation.
Microsoft and Databricks are also working to extend agent registry integration to Agent Bricks agents and tools. That is a direction for the joint work, distinct from the Genie agent registry synchronization described below.
Bring Genie agents into enterprise IT observability and security
As agents become part of everyday work, IT teams need visibility into the agent estate as well as controls over the data those agents use.
Genie agents sync into the Microsoft 365 agent registry, helping bring them into the organization's agent inventory. Agent 365 provides capabilities to register, observe, and govern agents across platforms, working with Microsoft Entra, Microsoft Purview, and Microsoft Defender.
Enterprise agent oversight complements Databricks governance in Unity Catalog and Unity Gateway. These layers address different responsibilities: understanding and managing agents across platforms, and governing data and AI assets within Databricks.
Microsoft Entra provides a shared identity foundation for the Azure Databricks experience. Together, these capabilities connect the user-facing AI experience with the identity, data permissions, and oversight that enterprise teams need to manage.
A shared foundation, useful across roles
Consider a business review supported by Copilot. A business user asks about performance. The data team's metric definitions give the question meaning. Copilot calls Genie for the relevant Databricks answer. Data permissions govern what can be accessed, while agent inventory and oversight help IT manage the agents involved.
That is the unified Azure Databricks story behind these announcements: make governed enterprise intelligence useful to more people, through the Microsoft experiences where their work happens.
Learn more at Microsoft Ignite
Databricks joins Ignite as a Visionary Sponsor, with six featured sessions, a sponsor booth, expert meet-ups, and dedicated customer conversations.
To explore the integrations and their technical context, start with these key sessions:
|
Session |
Title |
|
Powering AI with Azure Databricks, Foundry, Fabric, and Copilot | |
|
Build governed AI that understands your business context | |
|
Scale agentic AI and turn data into autonomous action |
The broader program also includes
BRK453: What's new and what's next in Azure Storage
THR909: Seismic shift: Turning subsurface data into decisions
LTG1303: Azure Databricks for government cloud: Modern data and AI, delivered
Find times, locations, and attendance options in the Microsoft Ignite session catalog.
Visit Databricks at booth 5271R and connect with the team through the Azure Expert Meet-Up programming for demos and technical conversations. Dedicated meeting space at The Social at the Hilton Canopy Hotel is available November 17-19.
Explore the Databricks Ignite program and meeting options or visit the Azure Databricks documentation to learn more about the platform.
Stay in the know with all things Azure Databricks https://aka.ms/adbreleasenotes