OpenAI rolls out finance-focused ChatGPT for banks and research teams

OpenAI has introduced a finance-specific version of ChatGPT aimed at financial institutions, combining premium data access, modelling support and enterprise security features in a single product. The offering, called ChatGPT for Financial Services, will initially target investment banking and equity research teams.

The company said its engagement with institutions such as Morgan Stanley and Evercore helped surface the main difficulties facing these teams. Reliable access to data and the creation of high-quality work products emerged as the most significant pain points, OpenAI said, adding that feedback from partners will inform future improvements and expansion into other financial services segments.

A central feature of the platform is built-in access to financial datasets from Daloopa, PitchBook, LSEG News and Crunchbase. The data spans earnings transcripts, financial statements, company fundamentals and information on private companies. OpenAI said teams can begin using these datasets without having to negotiate separate contracts or configure individual connectors. The company will index and host the data on its own infrastructure to improve retrieval speed and latency and to provide more detailed citations.

OpenAI described data as the foundation of all financial analysis, stressing the importance of dependable information for research and decision-making. The built-in datasets are intended to let professionals trace figures and claims back to their sources as they build their analysis. As an illustration, OpenAI said a banker working on a P&L normalisation exercise could examine the reconciliation and notes behind an adjusted EBITDA figure, see which costs were excluded and then determine how the number should be used in a valuation.

The product also provides access to OpenAI's latest frontier intelligence, with newer models to be made available as they are released. Beyond analysis, the platform addresses how financial teams turn their work into finished deliverables. Administrators can publish Excel, Word and PowerPoint templates through a dedicated admin page. Using firm templates and style guides, teams can produce valuation models, research notes and pitchbooks in their organisation's format.

On security, OpenAI said safeguarding material non-public information and client confidentiality is critical for financial institutions. The product builds on ChatGPT Enterprise controls, including SAML single sign-on, SCIM provisioning and role-based access controls. OpenAI said a firm's business data is not used to train its models by default, and that data is encrypted at rest and in transit, while administrators can configure workspace retention. Institutions can also manage access to skills and applications by role, enable or disable supported actions, and create multiple workspaces to enforce information barriers.

ChatGPT for Financial Services is available to eligible financial institutions, OpenAI said.