OpenAI Launches Data Agent in ChatGPT Work for Natural Language Data Analysis and Dashboard Creation

OpenAI has launched the Data agent in ChatGPT Work. This feature allows users to analyze company data, create interactive dashboards, and perform actions using only natural language questions, without the need to write queries or learn separate analysis tools.

Data Source Connectivity and Semantic Layer Utilization

The Data agent connects to major data warehouses such as Amazon Redshift, Google BigQuery, Snowflake, Databricks, MongoDB, and ClickHouse, as well as files from Google Drive and SharePoint. It also leverages context from trusted sources like Databricks Genie Ontology, dbt, and Snowflake Horizon to accurately interpret organizational business terms and metric definitions.

BI Tool Integration and Security Management

It integrates with external BI tools such as Tableau, Power BI, Sigma, ThoughtSpot, Omni, and Oracle BI, allowing natural language task instructions within existing environments. Enterprise admins manage data connection permissions and role-based access controls, and existing table, row, and column restrictions on connected accounts are strictly applied during queries.

Real-World Usage and Internal Adoption

Internally at OpenAI, nearly all product teams are using this feature, and more than two-thirds of the GTM organization are utilizing it. Alpha program customers such as NTT DATA, Thermo Fisher Scientific, and ServiceTitan have adopted the Data agent for sales analysis, report error detection, and opportunity assessment. ServiceTitan, for instance, built a dashboard showing that AI sidekick users had a campaign start rate approximately three times higher than non-users, using this insight to simplify onboarding.