Bring your business context to ChatGPT, Claude, Microsoft Copilot Cowork and Grok Bot

Databricks announced Genie Ontology, an enterprise context layer that combines governed definitions with continuously learned organizational knowledge to provide AI agents with the business understanding required for accurate insights. Without this context, agents may perform random searches across data sources, increasing the risk of incorrect answers. Genie Ontology acts as a map of the business, supplying definitions, relationships, and source context to help agents interpret terms correctly.

New Databricks Genie plugins integrate with ChatGPT, Claude, Microsoft Copilot Cowork, and Grok Bot, allowing teams to maintain business context in Databricks while making it available across preferred AI agents. Frontier agents can now access Genie Ontology to retrieve governed data, write and execute SQL, and return answers with citations. For complex questions, Genie routes tasks to specialized domain agents, such as marketing or finance, ensuring domain-specific analysis.

For clients using the MCP Apps extension, Genie returns interactive visualizations and citations directly in the conversation, enabling users to view charts, sources, and definitions without leaving their workflow. Agents handle routine tasks independently, such as monitoring product launches or analyzing campaign performance, while Genie provides the necessary business context and certified metrics. This allows agents to perform multi-step workflows, like preparing client briefings or drafting budget adjustments, with accuracy and governance.

Genie plugins extend governed business context through the Genie MCP server, accessible to AI assistants, coding agents, and client interfaces. Organizations can enable the plugin in their agent workspace to start asking business questions or assigning tasks, with Genie managing data retrieval and governance. Genie Ontology serves as the governed context layer, allowing businesses to define terms once and apply them across all connected agents, ensuring consistency and trust in AI-driven decisions.