Genie in Slack is now Generally Available: Governed AI cowork, right where work happens
Databricks launched Genie in Slack, enabling governed AI responses and actions directly within Slack channels and DMs using existing Databricks permissions and data governance.
Genie in Slack is now generally available, allowing users to DM or @mention Genie in a channel to receive governed answers grounded in their business context without leaving Slack. The app integrates with Databricks’ governance framework, ensuring responses reflect the user’s permissions through Unity Catalog and connected source permissions. Answers include citations and can be shared in-thread for team visibility or kept private based on workspace admin settings.
Teams can collaborate on Genie’s responses in any channel where the app is added, with Genie replying directly in the thread to maintain conversational context. Admins can configure channels to use Genie One by default or pin a specific Genie Agent for domain-specific queries, such as product FAQs or employee benefits. Users can also work privately in DMs, asking questions, uploading files for analysis, and taking approved actions in other tools without switching applications.
Genie in Slack supports turning insights into action, enabling users to orchestrate workflows and review processes directly from their Slack workspace. The app adheres to the same governance standards as other Genie surfaces, with every question processed using the asker’s identity and permissions. Workspace admins can centrally govern AI usage through Unity Gateway, including MCP policies, cost controls, and monitoring of Genie’s actions and tool access.
To use Genie in Slack, a Databricks account admin must configure a context-based ingress policy for secure connectivity between Slack and the Databricks workspace. Users need to sign in to the Genie app from a DM once before receiving results. The service routing messages between Slack and Databricks does not store questions or answers, ensuring data privacy.