JPMorganChase still tops AI ranking as big-bank lead widens

• Key insight: JPMorganChase, Capital One and RBC are the AI leaders among banks, according to Evident AI’s analysis.
• Expert quote: JPMorganChase has “an enterprise-wide AI strategy and execution and it’s building infrastructure that gets AI into the hands of everyone across its 22 lines of business and 10 functions — that is no small thing.” — Alexandra Mousavizadeh, co-CEO and co-founder of Evident.
• Forward look: Small banks can catch up by putting pressure on their vendors to improve their AI capabilities, experts say.

For the fourth year in a row, JPMorganChase , Capital One and RBC came in first, second and third, respectively, in a ranking of banks by AI leadership.

Processing Content

Research firm Evident AI scores the companies on their AI talent, innovation, leadership and transparency.

JPMorganChase’s top spot is logical. It’s the biggest bank with the largest tech budget and staff — why wouldn’t it be the leader? But according to Evident’s researchers, the bank performs above and beyond its natural advantages of staff and budget size.

It has “an enterprise-wide AI strategy and execution and it’s building infrastructure that gets AI into the hands of everyone across its 22 lines of business and 10 functions — that is no small thing,” Alexandra Mousavizadeh, co-CEO and co-founder of Evident, told American Banker. JPMorganChase did not respond to a request for comment.

The bank has the benefit of being an early mover, she said. It’s invested heavily in technology and it’s “laser focused on workflow transformation every single day,” she said. It’s currently working to lower its cost of delivery, she said.

Capital One has had an outstanding AI platform architecture for years, and RBC is “absolutely relentless and super aggressive on their line of business rollout and their platform build and their infrastructure,” Mousavizadeh said.

Some experts would like to see the AI leaders apply their technology toward improved products and services for customers and a better experience for employees.

“JPMorgan, Capital One and RBC have built substantial capabilities,” Bradley Leimer, strategic advisor at Darrery Capital and former head of innovation at Santander, told American Banker. “To me, the question is whether this translates into a better financial life for their customers, not just coding efficiency, or quicker pitch decks. Helping consumers and small businesses understand their cash flow, evaluate the impact of multiple decisions is a different test from counting research papers, change in talent, or internal use cases.”

Wells Fargo ranked fifth this year, up slightly from sixth last year. Citi rose from twelfth to eighth. Bank of America rose from tenth to seventh. Morgan Stanley dropped slightly from fifth to ninth.

The banks that lag behind the top ten do occasional rollouts and use copilots, “but they’re not setting up for speed and scale,” Mousavizadeh said. “The whole story this year is speed and scale,” which depend on having the right platform architecture.

A nimble platform architecture lets a company use a small language model where that makes most sense and a frontier model like Fable or Astra where necessary.

“Dynamic model routing that picks the right model for any given task keeps your costs down,” Mousavizadeh said. “This is what the leading banks are building or have built.”

Standing still means falling behind in this environment, she noted. “If you’re standing still, you have banks that are barreling ahead of you.”

Building harnesses, keeping people

The leading banks have all been focusing this year on building “harnesses” for agentic AI – guardrails and controls.

“Without that control set up, you can’t really do agentic AI,” Mousavizadeh said. “It’s still in its early days, but they’ve come a long way in getting the harness and the controls around that.”

Evident found 21% growth in “enablement talent,” in other words, engineers who develop controls and management tools for AI.

“They’re realizing they need great systems thinkers and engineers,” Mousavizadeh said. “They’re doing more complex things. We haven’t seen software engineers being laid off at the banks at all.”

This is part of a broader trend in which AI investments have not led to layoffs, Mousavizadeh said.

“Those banks that are leaning in the most and spending the most on AI are the ones that are hiring the most,” she said. “We see absolutely zero impact on headcount. Literally zero.”

Instead of cutting costs by reducing headcount, AI-forward banks are onboarding corporate clients faster with the help of AI.

“If you get the onboarding time down from six months to six days, you’re immediately generating revenue,” Mousavizadeh said.

In investment banking, research is being done faster and products are being delivered quicker with AI.

“It’s much more of a growth story, so we’re seeing the banks tilt their focus on the KPIs like revenue per head, market share growth, better customer delivery on the retail side, time to onboard,” she said.

Competition

Though the largest U.S. banks have held their AI lead for four years, the market is competitive, including on the AI front, Mousavizadeh said.

“You’ve got your big banks, J.P. Morgan, Citi, Bank of America, and on the investment banking side there’s Goldman and Morgan Stanley, and then you’ve got the pressures from the hedge funds like Jane Street, which are becoming massive challengers,” she said.

Brazilian neobank Nubank and British neobank Revolut, which have both received conditional approval from the Office of the Comptroller of the Currency for de novo bank charters, can build their own large language models as well.

And frontier labs like OpenAI and Anthropic are also becoming competitors in that they’re offering financial tools.

Smaller banks will eventually benefit from the democratization of AI through their vendors and platform providers, Leimer said.

“But this gap will likely widen before it closes as the leaders move toward agentic workflows that require massive institutional data sets to train,” he said.

“Smaller banks are in a race to improve efficiency and relevance,” Leimer said. “They will never be able to out-spend these giant banks on R&D. Instead, they should focus on highly specific, high-friction customer pain points.”

Their ability to provide personalized human service is their best asset, he said.

“AI should be used to make those human interactions deeper and more informed, not replaced by a generic chatbot,” Leimer said.