OpenAI Data agent in ChatGPT Work launch

TL;DR

OpenAI introduced the Data agent in ChatGPT Work, a plugin that lets users connect to enterprise data sources, ask plain‑language questions, and receive answers, visual dashboards, and actionable workflows without writing queries or learning new analytics tools.

Connect to trusted enterprise data sources

The Data agent supports a wide range of approved data warehouses and services, including Amazon Redshift, Datadog, Google BigQuery, ClickHouse, Databricks, MongoDB, Snowflake, as well as file stores like Google Drive and SharePoint. It leverages an organization’s semantic layer—such as Databricks Genie Ontology, dbt, GitHub, Snowflake Horizon, and existing BI dashboards—to interpret business terms, metric definitions, custom calculations, and data relationships. Enterprise administrators retain control over which connections are available and enforce existing permission models (table, row, column level).

“With the Data agent in ChatGPT Work, users can connect directly to Amazon Redshift, making it as easy as asking a question to explore data, surface insights, and make faster decisions across their organization.” —Naresh Chainani, Director of Engineering, AWS

“Connecting ClickHouse to the Data agent in ChatGPT Work puts real‑time analytics in front of all business users making decisions big or small.” —Ryadh Dahimene, Director of Product Management, ClickHouse

“Thousands of organizations rely on Databricks Genie to deliver trusted answers on their enterprise data… we are thrilled to partner with OpenAI to make it easy for all ChatGPT users to tap into Genie’s data intelligence.” —Ken Wong, Senior Director, Product Management, Databricks

“Our customers have built a trusted foundation for enterprise data and context in Snowflake… the Data agent lets employees investigate business questions while OpenAI models bring intelligence to experiences like Snowflake CoCo and CoWork.” —Umesh Unnikrishnan, Head of Developer Experiences, Snowflake

“Our customers’ most important data lives in Atlas, updating in real time… the Data agent can reach that data directly, so anyone can ask a plain‑language question and get an answer grounded in what’s happening right now.” —Pablo Stern, Chief Product Officer, AI and Emerging Products, MongoDB

“G2’s trusted B2B software data with the Data agent helps teams understand buyer needs, competitive dynamics, and emerging trends.” —Godard Abel, CEO and co‑founder, G2

Turn questions into analysis, visualizations, and actions

Users can ask follow‑up questions to drill into findings and view the underlying evidence. The agent can generate interactive dashboards with built‑in visualizations that are editable, shareable, and refreshable. Brand guidelines can be applied to match an organization’s look and feel.

The Data agent integrates with major BI platforms—Omni, Oracle BI, Power BI, Sigma, Tableau, and ThoughtSpot—allowing users to create or modify dashboards in the tools they already use via natural language commands.

“Connecting Tableau with ChatGPT Work brings trusted business semantics into a place employees already work, so the answers they get are grounded in the same data model.” —Southard Jones, EVP & Chief Product Officer, Tableau

“Power BI helps people make better decisions with a trusted semantic layer… the Data agent lets users create Power BI dashboards simply by describing their business questions.” —Bogdan Crivat, Corporate Vice President, Microsoft Fabric

“Our customers use Sigma for revenue forecasts… bringing that into ChatGPT Work means the fastest place to ask a question is also the place where answers are grounded in semantics.” —Hassen Karaa, SVP of Product, Sigma

“Analytics should work where your teams work. With the Data agent, teams can build ThoughtSpot AI analytics in minutes using natural language.” —Bhargav Addala, SVP, Product Management, ThoughtSpot

The agent can also recommend next steps, identify stakeholders, and disseminate findings through Slack or email, executing approved actions via connected tools.

Internal use and development at OpenAI

OpenAI’s own product and GTM teams heavily use the Data agent for internal analytics—over two‑thirds of the GTM organization rely on it. The internal data team built shared business definitions, access rules, and safeguards for sensitive data to enable this usage. A LinkedIn post and webinar provide deeper insight into the internal workflow.

Early customer experiences from the Alpha program

Multiple organizations in the Alpha program report tangible benefits:

  • NTT DATA: non‑engineers built and updated dashboards with plain language, reducing licensing and effort barriers.
  • Thermo Fisher Scientific: teams identified supply‑base opportunities and improved preparation.
  • ServiceTitan: discovered that users of the Atlas AI sidekick launched campaigns three times more often, informing onboarding improvements.
  • Zipline: natural‑language queries surfaced insights that would have taken top analysts hours to uncover.
  • Empower: combined organizational and AI usage data to surface top use cases from free‑form survey responses.
  • Piston: leadership accessed custom views of sales funnels, ticket volumes, and spend without manual report generation.
  • Doeren Mayhew: marketing and finance teams built tailored dashboards in two days, accelerating insight sharing.
  • CookUnity: built a high‑season conversion dashboard by asking, cutting planning time.
  • Turing: operational metrics were explored, revealing drivers and reducing manual reporting.
  • micro1: rebuilt performance tracking dashboards in half an hour while catching errors.
  • Unit8: business development leaders used scorecards to match pipeline demand with delivery capacity.

Getting started with the Data agent

The Data agent appears as Data in the Plugins directory of ChatGPT Work. Administrators enable it via Workspace → Plugins and configure the required data‑source plugins (e.g., Databricks, Snowflake). After installation and any necessary account‑connection steps, users start a conversation with @Data and ask business questions directly.

Sample prompts to try

  • Diagnose a metric change: “Diagnose why weekly active users changed last week. Identify likely drivers, compare against prior periods, and recommend next checks.”
  • Design a KPI framework: “Design a KPI framework for this new product area with primary metrics, drivers, guardrails, targets, and data validation needs.”
  • Create a leadership readout: “Turn this month’s metrics into a leadership‑ready update with actuals, comparisons, drivers, caveats, and recommended actions.”

These prompts demonstrate the agent’s ability to perform root‑cause analysis, design measurement structures, and generate executive‑ready reports.

Implications for enterprise analytics

The Data agent lowers the barrier to data‑driven decision‑making by removing the need for SQL expertise or specialized BI training. By embedding analytics directly into conversational workflows, organizations can accelerate insight generation, democratize access to trusted data, and reduce reliance on bottlenecked reporting teams. The integration with existing BI tools ensures that existing governance and visualization standards are respected while extending natural‑language capabilities to a broader user base.

This article summarizes OpenAI’s official announcement of the Data agent for ChatGPT Work, published on September 10 2026.

Sources