认识一下 ChatGPT Work 中全新的 Data agent：只需提问，就能把公司数据转化为答案、交互式仪表盘和行动。
各行各业的人都有数据能够回答的问题。销售为什么放缓了？哪里的支出在上升？哪些问题会威胁到我们最大客户的续约，我们应该优先解决什么？想要得到这些答案，往往意味着要等一份报告，或者请别人来做分析。
我们推出全新的 ChatGPT Work 中的 Data agent ，帮助更多人自己回答这些问题。它会连接你的公司数据，调查发生了什么变化，并构建你可以分享的交互式仪表盘。在一次对话中直接指导并细化分析，无需编写查询语句，也无需学习新的分析工具。
连接你的业务所信任的公司数据和上下文
Data agent 可连接已获批准的数据源，包括 Amazon Redshift、Datadog、Google BigQuery、ClickHouse、Databricks、MongoDB、Snowflake 等。它还能将 Google Drive 和 SharePoint 中的文件和文档纳入分析。
它使用你所在组织的业务术语、指标定义、自定义计算和数据关系来解读数据。这些上下文来自语义层和可信来源，例如 Databricks Genie Ontology、dbt、GitHub、Snowflake Horizon 和 BI 仪表盘。
平台合作伙伴
“借助 ChatGPT Work 中的 Data agent，用户可以直接连接 Amazon Redshift，像提一个问题一样轻松地探索数据、发现洞察，并在整个组织内更快地做出决策。”
——Naresh Chainani，AWS 工程总监
“将 ClickHouse 连接到 ChatGPT Work 中的 Data agent，能让所有业务用户在做大大小小的决策时都能用上实时分析。”
——Ryadh Dahimene，ClickHouse 产品管理总监
“数千家组织依赖 Databricks Genie 在其企业数据上提供可信的答案。Genie 拥有丰富的上下文，能够弥合从数据到洞察之间的鸿沟。我们很高兴与 OpenAI 合作，让所有 ChatGPT 用户都能轻松利用 Genie 的数据智能。”
——Ken Wong，Databricks 产品管理高级总监
“我们的客户已在 Snowflake 中为企业数据和上下文构建了可信的基础。借助 ChatGPT Work 中的 Data agent，员工可以利用这些数据和访问控制来探究业务问题——同时 OpenAI 模型也为 Snowflake CoCo 和 CoWork 等体验注入智能。无论团队选择在哪里工作，都能对业务拥有一致且受治理的理解。”
——Umesh Unnikrishnan，Snowflake 开发者体验负责人
“我们客户最重要的数据都存放在 Atlas 中，随着业务运转实时更新——每一笔订单、每一次会话、每一个事件都在发生时即刻落地。现在，ChatGPT Work 中的 Data agent 可以直接访问这些数据，因此人们无需再等待基于昨日数据生成的报告，而是可以用自然语言提问，并得到以当下正在发生的情况为依据的答案。”
——Pablo Stern，MongoDB 人工智能与新兴产品首席产品官
“G2 值得信赖的 B2B 软件数据与 ChatGPT Work 中的 Data agent 相结合，帮助团队理解买家需求、竞争动态和新兴趋势。它将来自经过验证的客户评价、买家意向信号和市场情报的洞察，直接带入团队做决策的工作流程中。”
——Godard Abel，G2 首席执行官兼联合创始人
企业管理员可以选择哪些数据连接可用，以及哪些角色可以使用它们。查询会强制执行所连接账户的现有权限，包括表、行和列级别的限制。
把问题转化为分析与行动
提出后续问题以深入探究结果，并查看每项发现背后的证据。
将分析转化为带有内置可视化的交互式仪表板。你的团队可以按需编辑、分享和刷新它。分享你的品牌指南，让输出契合你所在组织的视觉风格。
Data agent 还可以在 Omni、Oracle BI、Power BI、Sigma、Tableau 和 ThoughtSpot 中构建仪表板并与之交互。用自然语言在团队已经在使用的工具中直接下达工作指令。
BI 工具
“将 Tableau 与 ChatGPT Work 连接，能把可信的业务语义带入员工已经日常使用的场所，因此他们得到的答案与团队所依赖的数据模型保持一致。用户可以通过提问、探究证据，并借助内置的设计与分析最佳实践，将新的视图直接发布到 Tableau，从而轻松将洞察转化为行动。”
—Southard Jones，Tableau 执行副总裁兼首席产品官
“Power BI 通过数十万家企业使用的可信语义层，帮助人们做出更好的决策。借助 ChatGPT Work 中的 Data agent，用户有了一个新选择：只需描述自己的业务问题，就能创建 Power BI 仪表板，让更多人更容易理解数据并据此采取行动。”
—Bogdan Crivat，Microsoft Fabric 企业副总裁
“我们的客户使用 Sigma 进行收入预测和运营规划，因为其背后的指标、工作簿和分析都受到治理并可供共享。将其引入 ChatGPT Work，意味着提问最快的地方，如今也是答案以语义为基础的地方。”
—Hassen Karaa ，Sigma 产品高级副总裁
“分析应当在你团队工作的地方发挥作用。借助 ChatGPT Work 中的 Data agent，团队可以用自然语言在几分钟内构建 ThoughtSpot AI 分析。业务用户能够基于他们本就信任的定义获得受治理的答案，而开发者只需一个提示词就能交付可用的嵌入式分析应用。”
——Bhargav Addala，ThoughtSpot 产品管理高级副总裁
让 ChatGPT Work 推荐后续步骤并识别需要参与的人员。它可以通过 Slack 或电子邮件分享发现结果，并通过已连接的工具执行你批准的操作。
基于我们在 OpenAI 分析数据所用的工具构建
我们在 OpenAI 内部广泛使用 Data agent 背后的能力。我们几乎整个产品团队以及超过三分之二的 GTM 组织都在使用 ChatGPT Work 中的 data agent 自行分析公司数据。我们的数据团队通过创建共享业务定义、设置访问规则以及为敏感数据部署防护措施，使这一切成为可能。阅读这篇 文章 并参加我们的 网络研讨会 以了解更多。
NTT Data、Thermo Fisher、ServicePiston 以及我们 Alpha 计划中的其他组织正在使用 ChatGPT Work 中的 Data agent 来分析销售和支出、发现报告错误，并决定要追求哪些机会以及如何为其配置人员。
客户案例
“在 NTT DATA，许可成本、投入精力以及技术专长一直让我们难以在整个组织内扩大仪表盘的使用。借助 ChatGPT Work 中的 Data agent，许多非工程人员，尤其是销售和职能部门的人员，已经能够用自然语言构建和更新自己的仪表盘。”
——Yuji Shono，NTT DATA Group 全球 AI 办公室负责人
“在 Thermo Fisher Scientific，我们正在使用 ChatGPT Work 中的 Data agent，结合现有环境中的数据，帮助我们的团队更有效地做好准备，并发现与供应体系相关的机会。我们看到了更清晰的分析和更明智决策的潜力，这能帮助我们更好地服务客户。”
——Bharadwaj Ramachandran，AI 解决方案高级经理
“在 ServiceTitan，我们使用 ChatGPT Work 中的 Data agent 构建了一个仪表盘，显示 Atlas（我们的 AI 助手）的用户发起活动的频率大约是非用户的 3 倍。这一发现正在帮助我们简化入门流程，并帮助更多客户采用 Atlas。”
——Ankur Bhatt，ServiceTitan Agent OS 与应用 AI 工程副总裁
“在 Zipline，我们正在使用 ChatGPT Work 中的 Data agent 来探索公司数据，并仅通过自然语言提问就能构建仪表盘。在早期测试中，它揭示了一些深刻的发现，而如果由我们最优秀的人员之一去挖掘数据，可能要花上数小时才能发现。”
——Ryan Oksenhorn，Zipline 联合创始人
“在 Empower，我们使用 ChatGPT Work 中的 Data agent 整合组织数据和 AI 使用数据，发现采用趋势，并开展自己的分析。让我们印象深刻的是，它能如此轻松地理解自由形式的调查回复，找出最重要的用例和洞察，而如果靠人工完成，这些工作需要投入大量精力才能挖掘出来。”
——Tori Langlois，Empower 创新实验室赋能负责人
“我们的领导团队使用 ChatGPT Work 中的 Data Agent 更好地了解业务的方方面面——从分析销售漏斗和 Intercom 工单量，到审查 Brex 卡支出。每位领导者都能快速构建自定义视图，并深入挖掘最重要的趋势。”
——Shivam Shah，Piston 联合创始人兼 COO
“短短两天内，营销和财务等不同团队就开始使用 ChatGPT Work 中的 Data agent，将原始数据转化为针对各办公室量身定制的交互式仪表盘。它让复杂的报表变得更加易用，帮助我们向公司各层领导提供清晰、可分享的洞察。”
——Paul Griz，Doeren Mayhew 信息技术高级经理
“在 CookUnity，我们的增长团队使用 ChatGPT Work 中的 Data agent，仅通过提问就构建并完善了一个旺季转化仪表盘。我们将结果与内部报告进行了核对，并缩短了规划季节性获客支出所需的时间。”
——Dave Senior，CookUnity 应用 AI 负责人
“在 Turing，我们的业务运营团队使用 ChatGPT Work 中的 Data agent 来探索运营指标并提出后续问题。它揭示了变化背后的驱动因素，帮助我们减少手动报告，并优先改进入职流程和客户满意度。”
—Usama Nadeem，Turing 高级数据科学家
“作为 OpenAI 的评估合作伙伴，我们通过严格的评估，早已看到 ChatGPT Work 中 Data agent 的强劲表现。在 micro1 内部使用它，让我们有机会亲身感受到这一价值。我们的运营团队通过用自然语言提问，自行搭建了仪表盘，并在半小时内重建了我们的绩效跟踪仪表盘，同时还发现了原始版本中的错误。”
—Carlos Arroyo，micro1 战略项目负责人
“我们用它来了解各行业、各客户和各 AI 用例中需求最旺盛的地方，以及我们 pipeline 的变化如何与交付能力相匹配。我们的业务拓展和交付负责人可以编辑、分享和刷新这些记分卡，从而优先安排机会，并将合适的专业知识与客户需求相匹配。”
—Thanos Giannakopoulos，Unit8 AI&ML 技术总监
开始使用 Data agent
你会在 ChatGPT Work 的插件目录中看到 Data agent 以 Data 的形式列出。管理员可以通过 Workspace settings > Plugins 将其开放给团队使用或为团队安装。他们还可以启用并配置相关的数据源插件，例如 Databricks 和 Snowflake，并管理谁可以使用它们。
如果 Data 尚未安装，请在插件目录中找到它并选择 Install plugin，或直接前往 Data 列表页 。完成所需的账户连接步骤，然后与 @Data 开始对话并提出你的业务问题。
Meet the new Data agent in ChatGPT Work: turn your company’s data into answers, interactive dashboards, and action, just by asking.
People across every business have questions that data can answer. Why did sales slow down? Where is spending rising? Which issues threaten renewals in our largest accounts, and what should we fix first? Getting those answers often means waiting for a report or asking someone else to run the analysis.
We’re introducing a new Data agent in ChatGPT Work to help more people answer those questions themselves. It connects to your company data, investigates what changed, and builds interactive dashboards you can share. Direct and refine the analysis in one conversation, without writing queries or learning a new analytics tool.
Connect to the company data and context your business trusts
The Data agent connects to approved data sources including Amazon Redshift, Datadog, Google BigQuery, ClickHouse, Databricks, MongoDB, Snowflake, and more. It can also bring files and documents from Google Drive and SharePoint into the analysis.
It uses your organization’s business terms, metric definitions, custom calculations, and data relationships to interpret the data. This context comes from semantic layers and trusted sources such as Databricks Genie Ontology, dbt, GitHub, Snowflake Horizon, and BI dashboards.
Platform partners
“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. Genie has the rich context for bridging the gap from data to insights. 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. With the Data agent in ChatGPT Work, employees can tap into that data and access controls to investigate business questions—while OpenAI models simultaneously bring intelligence to experiences like Snowflake CoCo and CoWork. Teams share a common, governed understanding of their business wherever they choose to work.”
—Umesh Unnikrishnan, Head of Developer Experiences, Snowflake
“Our customers’ most important data lives in Atlas, updating in real time as their business runs—every order, every session, every event landing as it happens. Now the Data agent in ChatGPT Work can reach that data directly, so instead of waiting on a report built from yesterday’s numbers, 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 in ChatGPT Work helps teams understand buyer needs, competitive dynamics, and emerging trends. It brings insights from verified customer reviews, buyer intent signals, and market intelligence directly into the workflows where teams make decisions.”
—Godard Abel, CEO and co-founder, G2
Enterprise administrators choose which data connections are available and which roles can use them. Queries enforce the connected account’s existing permissions, including table, row, and column restrictions.
Turn questions into analysis and action
Ask follow-up questions to investigate the results and review the evidence behind each finding.
Turn the analysis into an interactive dashboard with built-in visualizations. Your team can edit, share, and refresh it as needed. Share your brand guidelines to tailor outputs to your organization’s look and feel.
The Data agent can also build and interact with dashboards in Omni, Oracle BI, Power BI, Sigma, Tableau, and ThoughtSpot. Direct the work in plain language in the tools your team already uses.
BI tools
“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 their teams rely on. Users can easily transform insights into action by asking questions, exploring evidence, and publishing new views directly to Tableau using built-in design and analytics best practices.”
—Southard Jones, EVP & Chief Product Officer, Tableau
“Power BI helps people make better decisions with a trusted semantic layer used by hundreds of thousands of enterprises. With the Data agent in ChatGPT Work, users have a new option to create Power BI dashboards by simply describing their business questions, making it easier for more people to understand their data and act on it.”
—Bogdan Crivat, Corporate Vice President, Microsoft Fabric
“Our customers use Sigma for revenue forecasts and operational planning because the metrics, workbooks, and analysis behind them are governed and shared. Bringing that into ChatGPT Work means the fastest place to ask a question is now 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 in ChatGPT Work, teams can build ThoughtSpot AI analytics in minutes using natural language. Business users get governed answers based on definitions they already trust, while developers ship working embedded analytics apps from a single prompt.”
—Bhargav Addala, SVP, Product Management, ThoughtSpot
Ask ChatGPT Work to recommend next steps and identify who needs to be involved. It can share the findings through Slack or email and carry out the actions you approve through connected tools.
Built from the tools we use to analyze data at OpenAI
We use the capabilities behind the Data agent broadly across OpenAI. Nearly all of our product team and over two-thirds of our GTM organization use data agents in ChatGPT Work to analyze company data themselves. Our data team made this possible by creating shared business definitions, setting access rules, and putting safeguards in place for sensitive data. Learn more by reading this post and attending our webinar .
NTT Data, Thermo Fisher, ServicePiston, and other organizations in our Alpha program are using the Data agent in ChatGPT Work to analyze sales and spending, catch reporting errors, and decide which opportunities to pursue and how to staff them.
Customer examples
“At NTT DATA, licensing costs, effort, and technical expertise have made it difficult to expand the use of dashboards across our organization. With the Data agent in ChatGPT Work, many non-engineers, particularly in sales and corporate functions, have been able to build and update their own dashboards using plain language.”
—Yuji Shono, Head of Global AI Office, NTT DATA Group
“At Thermo Fisher Scientific, we’re using the Data agent in ChatGPT Work with data from our existing environment to help our teams prepare more effectively and to identify opportunities with our supply base. We see potential for clearer analysis and better-informed decisions that can help us better serve customers.”
—Bharadwaj Ramachandran, Senior Manager AI Solutions
“At ServiceTitan, we used the Data agent in ChatGPT Work to build a dashboard showing that users of Atlas, our AI sidekick, launched campaigns at roughly three times the rate of nonusers. That finding is helping us simplify onboarding and help more customers adopt Atlas.”
—Ankur Bhatt, VP Engineering, Agent OS and Applied AI, ServiceTitan
“At Zipline, we’re using the Data agent in ChatGPT Work to explore our company data and build dashboards just by asking in natural language. In early testing, it surfaced profound findings that would have taken one of our best people hours of digging through the data to uncover.”
—Ryan Oksenhorn, Co-founder, Zipline
“At Empower, we used the Data agent in ChatGPT Work to combine organizational and AI usage data, uncover adoption trends, and do our own analysis. What impressed us was how easily it made sense of free-form survey responses, surfacing top use cases and insights that would otherwise have taken significant manual effort to uncover.”
—Tori Langlois, Innovation Lab Enablement Lead, Empower
“Our leadership team used the Data Agent in ChatGPT Work to better understand every part of our business—from analyzing the sales funnel and Intercom ticket volume to reviewing Brex card spend. Each leader could quickly build custom views and dig deeper into the trends that mattered most.”
—Shivam Shah, Co-founder and COO, Piston
“Within two days, different teams across marketing and finance were using the Data agent in ChatGPT Work to turn raw data into interactive dashboards tailored to each office. It’s made sophisticated reporting more accessible, helping us deliver clear, shareable insights to leaders across the firm.”
—Paul Griz, Senior Manager, Information Technology, Doeren Mayhew
“At CookUnity, our Growth team used the Data agent in ChatGPT Work to build and refine a high-season conversion dashboard just by asking. We checked the results against internal reports and cut the time needed to plan our seasonal acquisition spending.”
—Dave Senior, Head of Applied AI, CookUnity
“At Turing, our Business Operations team used the Data agent in ChatGPT Work to explore operational metrics and ask follow-up questions. It surfaced drivers behind changes, helping us reduce manual reporting and prioritize improvements to onboarding and customer satisfaction.”
—Usama Nadeem, Senior Data Scientist, Turing
“As an OpenAI evaluation partner, we’d already seen strong results from the Data agent in ChatGPT Work through our robust evaluations. Using it internally at micro1 gave us the chance to see that value firsthand. Our operations team built its own dashboards by asking questions in plain language, and in half an hour, rebuilt our performance tracking dashboards while catching errors in the original.”
—Carlos Arroyo, Strategic Project Lead at micro1
“We use it to understand where demand is strongest across industries, accounts, and AI use cases, and how changes in our pipeline align with delivery capacity. Our business development and delivery leaders can edit, share, and refresh those scorecards to prioritize opportunities and match the right expertise to client needs.”
—Thanos Giannakopoulos, Tech Director AI&ML, Unit8
Get started with the Data agent
You’ll find the Data agent listed as Data in the Plugins directory in ChatGPT Work. Administrators can make it available or install it for their teams through Workspace settings > Plugins. They can also enable and configure the relevant data-source plugins, such as Databricks and Snowflake, and manage who can use them.
If Data isn’t already installed, find it in the Plugins directory and select Install plugin, or go directly to the Data listing . Complete any required account-connection steps, then start a conversation with @Data and ask your business question.