For years, banks have sold investors on artificial intelligence (AI) with usage counts and hours saved. Few have said how much money it makes; at least, not on an earnings call.
That is starting to change. Bank of America has 130 to 140 AI projects running “at a cost of $400 million, generating a benefit of $800 million,” CEO Brian Moynihan said at a Barclays conference on Sept. 14, Fortune reported. He did not say what period the figures cover.
Bank of America will also double its AI expense budget next year, Moynihan said. The bank spends $4 billion a year on new technology initiatives, including AI, out of a $13.5 billion total technology budget, PYMNTS reported.
The returns show up most clearly in customer service. Erica, the bank’s virtual assistant, had 24.6 million active users in the second quarter, PYMNTS reported. Without it, the bank would need 11,000 more people to answer calls, Chief Technology and Information Officer Hari Gopalkrishnan told Fortune. EricaAssist, a generative AI tool used by more than 18,000 of the bank’s customer service representatives, cuts average call times by nearly a minute, PYMNTS reported.
The bank is also adding use cases behind the scenes. In July, Chief Financial Officer Alastair Borthwick said Bank of America had approved about 300 AI use cases and deployed more than 100 throughout the bank, PYMNTS reported.
We’d love to be your preferred source for news.
Please add us to your preferred sources list so our news, data and interviews show up in your feed. Thanks!
Société Générale and Commerzbank Put AI Savings in Their Targets
Société Générale tied its AI figure to a broader cost plan. The French bank expects about 1.9 billion euros (about $2.1 billion) in gross savings between 2026 and 2029, it said in its Sept. 21 strategic plan. AI accounts for up to 600 million euros (about $669 million) of that, according to the bank’s strategy presentation. The bank is targeting a cost-to-income ratio below 55% in 2029. It also signed a strategic agreement with Anthropic to speed up AI adoption across the company.
Commerzbank set out its AI goal in May. The German lender plans to invest about 600 million euros (about $669 million) in AI from 2026 to 2030, it said in its Momentum 2030 strategy update. AI should free up and partly redeploy about 10% of staff capacity, giving its sales teams more time to advise clients.
CEO Bettina Orlopp told shareholders at the bank’s annual meeting on May 20 that AI is “a key driver” of the bank’s transformation, the bank said. Orlopp said the AI investment should save about 350 million euros (about $390 million) a year by 2030, Bloomberg reported. That puts cost savings at about 70% of the 500 million euros (about $557 million) in yearly AI value the bank said in May it expects from 2030. Commerzbank is targeting a cost-to-income ratio of 43% by 2030.
90% of Large Firms See Some Returns From AI
Beyond banking, between 90% and 100% of large U.S. enterprises report some kind of return on AI spending, but only 5% to 10% report a full payback. That’s according to the PYMNTS Intelligence report “The Enterprise AI Payback Curve: Adoption Accelerates as Returns Take Shape,” which surveyed 60 senior technology executives at U.S. companies with at least $1 billion in revenue. At least half of firms in every industry group expect full payback to take five to six years.
Finance chiefs are measuring AI in financial terms, too. Among CFOs at U.S. companies with at least $1 billion in revenue, 75% measure generative AI’s returns by improved margins and 60% by lower operating costs. The share tracking higher revenue per customer jumped to 56.7% from 18.3% in July 2025, according to the PYMNTS Intelligence report “What Happens When CFOs Get Serious About Gen AI.”
For all PYMNTS AI coverage, subscribe to the daily AI Newsletter.
Add as Preferred Source
