OpenAI Launches ChatGPT for Financial Services, Targeting Bank Research and Pitch Decks
The tailored workplace product combines financial-data access and review controls, putting automation closer to work that has long trained junior bankers.
Listen to this story
The audio brief
Story brief
3 key pointsOpenAI is positioning finance as an early proving ground for industry-specific enterprise AI, offering a ChatGPT Work variant that can turn subscribed financial data into analyses and bank-formatted pitch decks. Morgan Stanley and Evercore helped design the product, which runs on GPT-6 Astra and includes citations, chart checks, and controls for sensitive deal materials. The commercial traction remains unclear:...
- 01
Native connections include LSEG, Daloopa, and PitchBook, alongside customers’ existing data subscriptions.
- 02
A demonstrated acquisition analysis and PowerPoint workflow showed intended use, not independent evidence of accuracy or performance.
- 03
OpenAI frames the product as increasing productivity per employee, while critics warn automation could weaken junior-bankers’ apprenticeship.
OpenAI has launched ChatGPT for Financial Services, a version of its workplace product built for investment banking and equity research. The system is designed to research companies, analyze financial data and generate presentation decks—work long associated with Wall Street’s junior bankers.
A banking-focused layer on ChatGPT Work
The product is a tailored version of ChatGPT Work, OpenAI’s enterprise offering. OpenAI said Morgan Stanley and Evercore participated as design partners. It uses GPT-6 Astra, OpenAI’s model for the release.
The immediate target is work done by investment-banking analysts and associates: researching deals, building financial analysis and producing pitchbooks. That focus distinguishes the product from a general workplace chatbot, even though its underlying platform is ChatGPT Work.
From subscribed data to a formatted deck
ChatGPT for Financial Services has native access to LSEG, Daloopa and PitchBook, and can connect to users’ existing data subscriptions. Those connections are meant to supply financial statements, earnings transcripts and other inputs to the system’s analysis.
In OpenAI’s demonstration, the platform assessed a potential acquisition target, pulled financial figures and produced a formatted PowerPoint deck based on a bank’s style guide. The demonstration showed the intended workflow; it was not an independent performance evaluation.
Controls intended for sensitive work
- Citations let users trace information back to source filings.
- Chart-auditing tools let users check charts against underlying data.
- Administrative controls are designed for sensitive deal materials.
Faster output, and a training question
Asked whether the release could reduce the need for junior bankers, Turley framed it as a way to increase productivity per employee. He compared its potential effect with Excel, which he said helped the industry produce better analysis faster.
That efficiency case sits beside an apprenticeship concern. Goldman Sachs partner Chris Churchman warned last month that automating tasks used to train junior bankers risks “cognitive atrophy” when reasoning is delegated away. The release does not settle whether banks will use the tool to reshape hiring, training, or both.
OpenAI said demand was strong but did not identify banks that had signed up. Turley said the company plans tailored solutions for additional sectors, making finance an early test of its industry-specific product strategy.
Sources
- cnbc.comOpenAI targets work of Wall Street junior bankers with new ChatGPT for Financial Services
Loading discussion...
Reader comments
Newest comments first. Replies stay oldest first.