OpenAI adds a Data agent to ChatGPT Work

OpenAI has introduced a Data agent in ChatGPT Work, available as a plugin that connects to approved company data sources and answers business questions without the user writing queries.

Per OpenAI's post, the agent reads from Amazon Redshift, Datadog, Google BigQuery, ClickHouse, Databricks, MongoDB and Snowflake, and can pull files from Google Drive and SharePoint into an analysis. It draws metric definitions, custom calculations and data relationships from semantic layers and sources including Databricks Genie Ontology, dbt, GitHub, Snowflake Horizon and existing BI dashboards. Beyond producing its own interactive dashboards, it can build and manipulate dashboards in Omni, Oracle BI, Power BI, Sigma, Tableau and ThoughtSpot, and push findings to Slack or email.

Access is administrator-gated: enterprise admins pick which connections are available and which roles can use them, and queries run under the connected account's existing table, row and column permissions. Users invoke it with @Data after installing it from the Plugins directory in Workspace settings.

OpenAI says nearly all of its product team and more than two-thirds of its go-to-market organization use data agents in ChatGPT Work internally. NTT Data, Thermo Fisher and ServicePiston are named among Alpha program participants using it for sales and spending analysis and catching reporting errors.

The company did not publish accuracy benchmarks, latency figures, pricing, or a timeline for general availability beyond the Alpha program, and gave no measured results from the named customers.