An analytics agent still needs a shared definition of the business

OpenAI’s September release connects to approved sources including Redshift, BigQuery, ClickHouse, Databricks, MongoDB, and Snowflake. It can also incorporate documents from services such as Drive and SharePoint.

The consequential ingredient is the business context: metric definitions, custom calculations, terminology, and relationships between datasets. The product draws on semantic layers and trusted sources including dbt, Databricks Genie Ontology, and Snowflake Horizon.

Administrators select connections and permitted roles. Queries use the connected account’s existing restrictions, including controls at table, row, and column level.

The agent can produce interactive dashboards and work with established BI tools such as Tableau and Power BI. Users can follow up on findings and inspect the supporting evidence.

OpenAI says nearly all its product organization and more than two-thirds of its go-to-market organization use the underlying capabilities internally. Its data team prepared that deployment by establishing shared business definitions, access rules, and safeguards for sensitive information.