OpenAI has introduced a new Data agent in ChatGPT Work, designed to help individuals analyze business data through natural language queries. This agent connects to various company data sources to provide answers, generate interactive dashboards, and facilitate data-driven actions.
The Data agent allows users to investigate changes, build shareable dashboards, and refine analyses within a single conversation, eliminating the need for manual queries or specialized analytics tools.
Key Points
- The Data agent in ChatGPT Work connects to approved data sources such as Amazon Redshift, Datadog, Google BigQuery, ClickHouse, Databricks, MongoDB, and Snowflake.
- It can also incorporate files and documents from Google Drive and SharePoint into its analysis.
- The agent utilizes an organization's business terms, metric definitions, custom calculations, and data relationships from semantic layers and trusted sources like Databricks Genie Ontology, dbt, GitHub, Snowflake Horizon, and BI dashboards.
- Enterprise administrators control available data connections and user roles, with queries enforcing existing permissions including table, row, and column restrictions.
- The Data agent can build and interact with dashboards in Omni, Oracle BI, Power BI, Sigma, Tableau, and ThoughtSpot.
- It can recommend next steps, identify involved parties, and share findings via Slack or email, executing approved actions through connected tools.
- NTT Data, Thermo Fisher, and ServicePiston are among the organizations that participated in the Alpha program for the Data agent.
Context
According to OpenAI, the Data agent addresses the common challenge of waiting for reports or requesting analysis from others to answer business questions. The agent is built upon capabilities used broadly within OpenAI, with nearly all of their product team and over two-thirds of their GTM organization utilizing data agents in ChatGPT Work for internal data analysis.
Why It Matters
This development offers a way for businesses to empower more employees to perform self-service data analysis, potentially reducing reliance on specialized data teams for routine inquiries and accelerating insight generation.
What To Do
- Check the Plugins directory in ChatGPT Work for the Data agent listed as Data.
- Administrators can make the Data agent available or install it for teams via Workspace settings > Plugins.
- Enable and configure relevant data-source plugins, such as Databricks and Snowflake, and manage user access.
- Begin a conversation with
@Datain ChatGPT Work to ask business questions or request analysis.
