Artificial intelligence commercialization in vertical industries is advancing deeply. On September 10, OpenAI officially announced the launch of a customized service for financial services teams—ChatGPT for Financial Services. Developed jointly by OpenAI, Morgan Stanley, and Evercore, this product initially focuses on investment banking and stock research scenarios, marking the beginning of large model technology truly penetrating the core operations of financial institutions.

In terms of core technical architecture, the service integrates the latest GPT-6 Astra inference capabilities, enabling efficient processing of numbers, tables, and footnotes in massive financial documents. With this powerful multimodal and logical analysis capability, financial teams can comprehensively research information across multiple data sources, track data changes over different periods, generate interactive charts, and directly conduct complex financial modeling, write research reports, and create customized client pitch materials.

To address the high requirements of the financial industry for data accuracy and compliance, the product includes authoritative financial data from mainstream providers such as Daloopa, PitchBook, LSEG News, and Crunchbase, covering earnings calls, financial statements, company fundamentals, and private company information. All relevant data is indexed and hosted by OpenAI, allowing users to make fine-grained citations and trace specific data and conclusions at any time. In addition, OpenAI plans to collaborate with organizations such as S&P Global, LSEG, MSCI, Dow Jones Factiva, and Moody's to support users in accessing authorized content through existing financial data subscriptions, and has optimized the MCP connectors of financial data providers, fully integrating more than 50 connectors including Datasite, Box, Preqin, and Intapp.

In terms of enterprise-level customization and application implementation, financial institutions can independently publish templates such as Excel, Word, and PowerPoint through a dedicated management page. After administrators set up enterprise templates and format specifications, team members can quickly convert analysis results into valuation models and professional documents that meet institutional standards. At the same time, financial institutions and their developers can directly build customized applications tailored to specific financial business through APIs.

In terms of enterprise data security and compliance management, the product inherits the security standards of ChatGPT Enterprise, supporting SAML single sign-on, SCIM user provisioning, and role-based access control. OpenAI explicitly commits that enterprise business data will not be used for model training by default, and data will be encrypted during transmission and static storage. Administrators can flexibly set the specific data retention period for workspaces. In addition, compliance teams can export workspace logs via the OpenAI Compliance Platform to meet audit and investigation needs. Enterprises can also finely manage reading and writing permissions for skills and applications by role, and strictly implement information isolation by creating multiple workspaces. Currently, the product is officially available to qualified financial institutions.