Databricks AI-powered features
This page introduces AI-powered features in Databricks that can help you work more efficiently. These include Genie for natural language data exploration, Databricks Assistant for coding and dashboard building, and auto-generated table documentation. Access to Databricks AI-powered features is controlled by the partner-powered AI assistive features setting, which is enabled by default.
What are Databricks AI-powered features?
Databricks AI-powered features use a compound AI system that combines the use of AI models, retrieval, ranking, and personalization systems to understand your organization’s data and usage patterns. These features support tools such as Databricks Assistant, AI-generated comments, and intelligent search, improving accuracy and relevance in results across the workspace. These Databricks AI-powered features enable everyone in an organization to be more productive using data and AI, while maintaining the governance and controls established in Unity Catalog.
When the partner-powered AI assistive features setting is turned Off, Databricks Assistant uses a Databricks-hosted model. Features, such as Genie, that rely on external models are not available.
Databricks AI features: Trust and safety
Databricks understands the importance of your data and the trust you place in us when you use Databricks services. Databricks is committed to the highest standards of data protection and has implemented rigorous measures to ensure your information is protected. For more details, see Databricks AI features trust and safety.
How do I enable or disable Databricks AI features?
Databricks AI-powered features are enabled by default. An admin can disable or enable Databricks AI-powered features for all workspaces in an account. If an admin has permitted workspace setting overrides, workspace admins can enable or disable Databricks AI-powered features for specific workspaces. For more details, see For an account: Disable or enable partner-powered AI features.
Use Databricks Assistant to develop code
Databricks Assistant works as an AI-based companion pair-programmer to make you more efficient as you create notebooks, queries, and files. It provides inline code suggestions as you type, can help you rapidly answer questions by generating, optimizing, completing, explaining, and fixing code and queries. Databricks Assistant is available in notebooks, the SQL editor, and when creating dashboards. For details, see Enable and try Databricks Assistant and What is Databricks Assistant?.
To learn about command shortcuts, see Command shortcuts for notebooks.
Enable natural language data exploration with Genie spaces
AI/BI Genie uses AI-powered features to allow business teams to interact with their data using natural language. It uses comments and metadata from your Unity Catalog data sources, along with user-provided instructions and example queries, to interpret questions and generate relevant SQL queries. Genie learns from context to better understand business logic, apply company-specific terminology, and return accurate results that support intuitive data exploration. See What is an AI/BI Genie space.
Build dashboards and enable user-driven analysis
When drafting a dashboard, you can use natural-language prompts to generate visualizations. For details, see Create visualizations with Databricks Assistant.
You can also automatically create a companion Genie space from a dashboard (Public Preview). Users can open the associated Genie space to ask follow-up questions and explore the data more deeply. See Enable a Genie space from your dashboard.
AI-generated table comments in Catalog Explorer
For tables in Catalog Explorer, Databricks automatically generates comments that describe the table based on the table metadata. You do not have to do anything to generate this documentation. You can edit the comment, accept it as-is, or delete it. For detail, see Add comments to data and AI assets.
Use Databricks Assistant for help
The help assistant is always available to you in the Databricks workspace. You can type in a question, and it generates answers based on Databricks documentation and knowledge base articles. For details, see Get help in the Databricks workspace.