From JPMorgan to Morgan Stanley: How Wall Street Banks Are Spending on AI

Generative AI is reshaping nearly every corner of Wall Street, from trading floors to back offices, from workflows to corporate culture. At the same time, analysts are increasingly pressing executives on whether AI spending is paying off in terms of real revenue and security. Business Insider looked at how major banks are rolling out artificial intelligence.

JPMorgan, the country's largest bank, is aiming for leadership in the AI race with a technology budget of nearly $20 billion a year. On the bank's second-quarter earnings call, CEO Jamie Dimon said JPMorgan has almost 1,000 AI use cases, spanning everything from fraud detection to marketing and drafting memos, and that the firm's in-house generative AI platform has been rolled out to more than 200,000 employees. In July, the bank's head of AI, Teresa Heitsenrether, announced her retirement after four decades in the industry, prompting a reshuffle of the tech leadership team.

Goldman Sachs has earmarked $6 billion for technology this year, but CEO David Solomon has said he'd like to push that to at least $8 billion, though he can't justify it because he has to deliver earnings. Citi, meanwhile, is trying to drive employee adoption through a mix of "metrics and pride," according to the bank's technology chief, Tim Ryan — CEO Jane Fraser has said nearly 90% of staff now use AI tools.

In July, Wells Fargo unveiled an "AI copilot" service designed to boost financial advisors' day-to-day productivity, and CEO Charles Scharf said generative AI tools have lifted engineers' productivity by as much as 35%. Bank of America, for its part, approved a $13 billion technology budget for 2025, greenlit more than 300 AI and machine-learning use cases, and has fully deployed at least 34 generative AI applications — its virtual assistant, Erica, has fielded more than 3.2 billion user interactions since 2018.

Morgan Stanley is leaning on its years-long partnership with OpenAI: from January through June, its DevGen.AI tool saved developers more than 280,000 hours, or the equivalent of 11,666 workdays. In a survey of the bank's summer interns, 72% said they use ChatGPT daily or several times a week — a clear signal of how the next generation of financiers views artificial intelligence.

Taken together, the picture shows Wall Street banks now treating AI not as an experiment but as a strategic priority — though each bank is charting its own course, from centralized oversight to efforts aimed at driving employee engagement.