Databricks Introduces Genie One MCP for Governed Business Context in AI Agents
Databricks has launched Genie One MCP, a new capability designed to provide governed business context from Genie Ontology to third-party AI assistants and coding agents. This offering addresses the critical problem of general-purpose AI agents lacking reliable interpretation of enterprise data, approved definitions, and access permissions. By operationalizing trusted business context, Genie One MCP aims to enable more reliable decisions and greater productivity for teams. It is currently available for preview, integrating with popular tools like ChatGPT, Claude, and Microsoft Copilot.
- →Genie One MCP Brings Governed Business Context to AI Agents
- →Extending Context to Third-Party AI Assistants and Coding Agents
- →Addressing the 'Context Problem' for AI Agents
- →Genie Ontology Provides Governed Context Layer
- →Real-world Use Cases for Genie One MCP
Features (2) ›
- Genie One MCP Brings Governed Business Context to AI Agents
Genie One MCP operationalizes trusted business context across the AI cowork tools teams already use, enabling more reliable decisions and greater productivity. It connects AI assistants like ChatGPT, Claude, and Microsoft Copilot to Genie One to answer business questions using approved definitions, trusted data relationships, and permission-aware access controls.
- Extending Context to Third-Party AI Assistants and Coding Agents
While Genie One provides a native AI cowork experience in Databricks, Genie One MCP extends the same governed business context to other approved AI assistants and coding agents. This allows organizations to define business meaning once in Genie Ontology and apply it consistently across their evolving AI ecosystem.
Enhancements (1) ›
- Genie Ontology Provides Governed Context Layer
Genie Ontology is Databricks' governed context layer that helps Genie One understand the business by combining approved metrics, business definitions, relationships, ownership, and access policies. It uses this context to identify correct definitions and sources, reconcile conflicts, and enforce user permissions, resulting in traceable, permission-aware answers.
Notes (3) ›
- Addressing the 'Context Problem' for AI Agents
General-purpose AI assistants often struggle with fragmented data, inconsistent definitions, and missing authority, leading to unreliable answers to complex business questions. Direct data access alone does not provide the shared business context needed to deliver governed responses, introducing challenges in accuracy, cost, governance, and consistency.
- Real-world Use Cases for Genie One MCP
Genie One MCP can enhance AI assistant capabilities for tasks such as campaign performance reviews for marketing leaders, monthly business reviews for finance leaders, and customer retention analyses for customer success teams. It connects relevant enterprise data and context, allowing AI agents to generate actionable insights and plans.
- Preview Availability and Setup
Genie One MCP is available via a preview toggle in a Databricks workspace, requiring a client connection specific to the AI assistant. Detailed setup instructions are provided in the Databricks AI assistants and coding agents documentation.
https://www.databricks.com/blog/genie-one-mcp-give-any-ai-agent-right-business-context
Related releases
- Databricks Launches Genie One MCP for Governed AI Agent Context Databricks Blog ·
- Databricks explains why AI analytics requires governed data for trustworthy insights Databricks Blog ·
- Data Ontology: Providing Business Context for Trustworthy AI Agents Databricks Blog ·
- Databricks SQL 2026.36 Rolling Out to Current Channel Databricks Release Notes ·
- Databricks Genie One MCP Server GA; Beta Endpoint Deprecated Databricks Release Notes ·
- Databricks SQL Version 2026.36 Now Available in Preview Channel Databricks Release Notes ·