AI in Tableau and Data Usage
AI in Tableau is built on data. As a Tableau customer, it’s important for you to know what data is being used in the features where you’ve enabled AI.
This table lists both the Customer Data, which is data submitted by the Customer to our services as defined in our Main Services Agreement (MSA)(Link opens in a new window), and the usage data, which is data relating to users’ interactions with Tableau.
Data submitted to a feature with AI in Tableau enabled may be used to improve your services and features or to develop new features that you can access without additional cost.
Note: The features in this list do not use global models, which are models that look for aggregated, anonymous trends across multiple Tableau Customers.
| Product | Feature | Data Used | Generative AI Model Used |
| Tableau Cloud | Enhanced Q&A (Discover) in Tableau Pulse Tableau Pulse insight summaries | The metric definition. This includes the measure, time dimension, aggregation, adjustable dimension filters, currency unit and custom unit name. Data the customer provides in the Published Data Sources that are used to create Tableau Pulse metrics. Data the customer provides when creating metric definitions in Tableau Pulse. Statistical insights derived from a combination of the above data. | Azure OpenAI |
| Tableau Cloud, Tableau Desktop, Tableau Server | Tableau Agent-assisted viz authoring | Questions the customer types in the Tableau Agent text box. Metadata descriptions of the data sources used with Tableau Agent Fields in the data source, including caption, description (used for viz generation only), role, field type, data type, default aggregation, calculation formulas, field table name and placements of fields in the viz. Data source metadata, such as ID or name, LUID and valid calculation functions. A sampling of data (1,000 unique values or less) from the data source used with Tableau Agent. | Tableau Cloud: Azure OpenAI Tableau Server: OpenAI |
| Tableau Cloud, Tableau Desktop, Tableau Server | Tableau Agent – AI-assisted colour palettes in viz authoring in Tableau Desktop | The colour palette description that the customer types into the text box. The colour palette type (Categorical, Diverging or Sequential). | Tableau Cloud: Azure Open AI Tableau Server: OpenAI |
| Tableau Cloud, Tableau Desktop | Tableau Agent for dashboards (Beta) – create dashboard summaries and insights | Dashboard Overview:
Dashboard Insights:
| |
| Tableau Cloud, Tableau Server, Tableau Prep Builder | Tableau Agent – flow assistant in Tableau Prep | Questions the customer types in the Tableau Agent text box. Feedback the customer submits through Tableau Agent. Metadata descriptions of the data sources and flow steps used with Tableau Agent (such as fields in the data source and placements of fields in the flow). Field names (not descriptions) and field data types A sampling of data from the data source used with Tableau Agent. No data is excluded from the data source. | Tableau Cloud: Azure OpenAI Tableau Server: OpenAI |
| Tableau Cloud, Tableau Server, Tableau Prep Builder | Tableau Agent – assisted calculation creation in Tableau Prep | Questions the customer types in the Tableau Agent text box. Metadata for the Tableau Prep flow step that includes the calculated field (such as field names and data types). | Tableau Cloud: Azure OpenAI Tableau Server: OpenAI |
| Tableau Cloud | Tableau Catalog - draft descriptions for data sources, workbooks and tables | Metadata descriptions of data assets (data sources, workbooks and tables). | Azure OpenAI |