OpenAI's new ChatGPT for Financial Services, built with Morgan Stanley and Evercore, automates the research, modeling, and pitchbook work that has long defined a Wall Street analyst's first years on the job.
Entry-level investment banking has run on the same model for decades: hire sharp new graduates, put them through roughly 100-hour weeks building pitchbooks and comparable-company analyses, and let the workload itself do the training. OpenAI's new ChatGPT for Financial Services is built to take over a meaningful share of that exact workload. Developed with Morgan Stanley and Evercore as design partners, the tool researches companies, builds valuation models, and drafts client-ready presentations, the same tasks that have traditionally defined an analyst's or associate's first few years on the job.
What changed
The product pairs OpenAI's GPT-6 Astra model with financial data built directly into the product: earnings transcripts, company fundamentals, and private-market data from providers including Daloopa, PitchBook, LSEG News, and Crunchbase. Firms that already subscribe to S&P Capital IQ, LSEG, MSCI, Dow Jones Factiva, or Moody's can carry over their existing entitlements through single sign-on rather than negotiating new contracts, and administrators can publish a firm's own PowerPoint, Word, and Excel templates so the output matches house style and formatting conventions.
According to CNBC, OpenAI vice president of product Nick Turley described the goal plainly during a press briefing: "We're effectively teaching ChatGPT to research like an analyst and back up its conclusions like an analyst as well." In a live demonstration, the platform analyzed a potential M&A target, pulled financial figures from industry-standard data sources, built a peer comparison, and generated a formatted PowerPoint deck in a bank's own template. OpenAI reports its model scores 69.9 percent on an internal benchmark it calls OfficeQA Pro, which tests finding and interpreting figures across Treasury Bulletins and financial tables, compared with 60.2 percent for its prior model, GPT-5.6 Sol.
The rollout also fits a broader strategic pattern. CNBC reports that OpenAI's finance chief, Sarah Friar, told investors in August that the company's enterprise business now brings in more revenue than its consumer business, the one that took off after ChatGPT's 2022 launch. Anthropic already sells a competing product, Claude for Financial Services, and Turley told reporters OpenAI plans similar tailored offerings for other industries beyond finance.
Why it matters
For a bank, the research-and-formatting workload this tool automates is not incidental. It's the training ground junior staff have used for decades to build the judgment that eventually makes them senior dealmakers. When CNBC asked Turley whether the product would reduce the need to hire junior bankers, he framed it as a productivity gain instead, comparing it to what Excel did for the industry decades earlier: "I think in the same way that Microsoft Excel transformed the industry and allowed them to produce better analysis faster, you will see technology like this do the same."
For a business leader outside finance, the more transferable detail is the shape of the substitution rather than the specific spreadsheets. A model is now handling multistep research-then-format work that used to require a person building institutional knowledge through repetition. That pattern shows up anywhere a junior role's core value was doing structured, high-volume work long enough to absorb the reasoning behind it. Compressing that work into minutes doesn't remove the training question, it just relocates it.
The honest caveat
OpenAI has not named a single bank that has signed on to pay for the product, and access is currently limited to eligible institutions who contact sales directly, so there is no independent data yet on how much work this actually removes versus reshapes. The tension it raises is real and unresolved. Chris Churchman, the Goldman Sachs partner leading one of the bank's flagship AI projects, warned last month that automating the tasks junior bankers use to learn the job risks what he called "cognitive atrophy" in the next generation of financiers, telling CNBC: "Reasoning is still important. You still need to reason about problems and structure it into an argument, and now we're delegating reasoning."
Closing observation
Excel didn't eliminate the analyst seat, it changed what analysts spent their extra hours doing. Whether this generation of tools follows that same path, or actually narrows the pipeline that produces the next generation of senior bankers, is a question the industry hasn't answered yet, and won't be able to until banks start saying which one of them signed up.