OpenAI has introduced a Data agent in ChatGPT Work designed to let employees query corporate data in natural language, get explanations of what changed, and create interactive dashboards and action plans without writing queries or learning a new analytics tool.
Why this matters
Many routine business questions — for example, why sales slowed, where spending is rising, or which issues threaten renewals for major accounts — require data to answer. Previously, obtaining those answers often meant waiting for a report or requesting analysis from someone else. The Data agent aims to broaden who can run those analyses directly.
Capabilities
- Connects to approved data sources including Amazon Redshift, Datadog, Google BigQuery, ClickHouse, Databricks, MongoDB, Snowflake, and can also bring in files and documents from Google Drive and SharePoint.
- Uses the organization’s business terms, metric definitions, custom calculations and data relationships to interpret results; this context is supplied by semantic layers and trusted sources such as Databricks Genie Ontology, dbt, GitHub, Snowflake Horizon and existing BI dashboards.
- Enforces existing access controls: enterprise administrators choose which data connections are available and which roles can use them; queries respect the connected account’s permissions, including table-, row- and column-level restrictions.
- Supports follow-up questions so users can probe findings and review the evidence behind each conclusion.
- Converts analysis into interactive dashboards with built-in visualizations that teams can edit, share and refresh. Outputs can be tailored to an organization’s brand guidelines.
Integrations and workflow
The Data agent can build and interact with dashboards in BI tools such as Omni, Oracle BI, Power BI, Sigma, Tableau and ThoughtSpot. Teams can direct work in the tools they already use using plain-language instructions.
The agent can recommend next steps, identify people to involve, share findings via Slack or email, and carry out approved actions through connected tools.
Internal adoption and safeguards
OpenAI reports that the Data agent is used broadly internally: nearly all of its product team and over two-thirds of its go-to-market organization use data agents in ChatGPT Work to analyze company data themselves. OpenAI’s data team created shared business definitions, set access rules, and put safeguards in place for sensitive data to enable this usage.
Alpha participants and use cases
Organizations in the Alpha program, including NTT Data, Thermo Fisher and ServicePiston, are using the Data agent to analyze sales and spending, detect reporting errors, and decide which opportunities to pursue and how to staff them.
Access and installation
In ChatGPT Work the agent appears as Data in the Plugins directory. Administrators can make it available or install it for teams via Workspace settings > Plugins, enable and configure relevant data-source plugins (such as Databricks and Snowflake), and manage who can use them. If Data is not already installed, users can find it in the Plugins directory and select Install plugin, complete any account-connection steps, then start a conversation with @Data and ask a business question.
Example prompts
- @Data Diagnose why weekly active users changed last week. Identify likely drivers, compare against prior periods, and recommend the next checks.
- @Data Design a KPI framework for this new product area with primary metrics, drivers, guardrails, targets, and data validation needs.
- @Data Turn this month’s metrics into a leadership-ready update with actuals, comparisons, drivers, caveats, and recommended actions.
Summary
The Data agent in ChatGPT Work enables natural-language interrogation of corporate systems, creates shareable, interactive dashboards from the results, and operates within existing organizational access rules. It relies on organizational context such as metric definitions and semantic layers to provide reliable analysis and supports integration with BI tools and collaboration channels for sharing and execution.
