How Agentic AI Will Transform Banking (and Banks)

The technology is developing at warp speed as it promises to increase efficiencies and reduce costs. That’s creating winners and losers in the banking industry.

By Lynnley Browning

Published on February 5th, 2025 in Artificial Intelligence

Reports of China’s low-cost AI model, DeepSeek, have stunned fintechs and chipmakers that make the hardware for artificial intelligence, but there’s another big story unfolding for US banks: agentic AI.

First, some definitions.

Agentic AI systems combine machine learning, large language models synonymous with Chat GPT maker OpenAI, and enterprise-wide automation. They’re fully autonomous and more adaptable to new inputs. Think of them as a non-human brain that parses reams of data for insights on functions including risk management and trading, strategizes and makes autonomous decisions about what to do with that information, and learns from the experience. IBM said last October that they’re "the next big thing in AI research." Bank of America wrote in January 2025 that agentic AI is developing so rapidly that over the next decade, it may ultimately "alter" bank operations reliant on human capital and "spark a corporate efficiency revolution that transforms the global economy."

Agentic AI vs. AI Agents

Agentic systems sometimes get conflated with AI agents, a closely related technology whose poster child is customer service chatbots. It’s helpful to regard agents as subset of an agentic system, able to execute a specific task set into motion by an agentic AI decision, such as putting a suspicious account on hold for review. Think of AI agents as the robotic arms of the non-human brain.

Read more:

Rik Reppe, the vice president of Acquis Cortico-X, a business-technology consulting firm, likens agentic AI to "the badass barista that knows every customer who walks in and that you haven’t gotten the delivery today so you don’t have any oat milk and your espresso machine only has one side of it working."

Meanwhile, he said, AI agents are "the espresso machine and the blender and the frother." As Cortico-X puts it, "Agentic AI orchestrates outcomes, while AI agents execute tightly defined tasks."

Why does the difference between the two matter? Because a bank that buys an AI product or service from the plethora of fintech software startups — the latter a $17 billion market last year that will more than quadruple by 2032 — needs to understand what it’s solving for with regard to levels of autonomy, goal orientation, learning capabilities and complexity.

What if a bank’s agentic AI (the barista) automatically instructs an AI agent (the frother) to make a trade (a skinny latte) that turns out to be loss making, or authorizes discriminatory lending decisions based on flawed data or assumptions? With agentic AI, "there are so many challenges relative to regulation and ethics and risk management," says Jim Perry, a senior strategist at bank consulting firm Market Insights.

As such, "it’s really only the largest banks that are doing any kind of experimentation with it."

Agentic AI’s Intertwining Vectors and the ‘Do it For Me’ Economy

Agentic AI has two intertwined vectors. For banks, one path is internal, and focused on operational efficiency for tasks including the automation of routine data entry and compliance and regulatory checks, summaries of email and reports, and the construction of predictive models for trading and risk management to bolster insights into market dynamics, fraud and credit and liquidity risk.

The other path is consumer facing, and revolves around managing customer relationships, from automated help desks staffed by chatbots to personalized investment portfolio recommendations. Both trajectories aim to improve efficiency and reduce costs. Agentic AI "could have a bigger impact on the economy and finance than the internet era," Citigroup wrote in a January 2025 report that calls the technology the "Do It For Me" Economy.

JPMorgan Chase, followed by Capital One and Royal Bank of Canada, lead Evident AI’s 2024 index for "AI maturity," a gauge of progress in talent (hiring AI developers and data and software engineers), leadership (C Suite and communications about AI), innovation (long-term investment in AI), and transparency focused on "responsible" and ethical AI that doesn’t trip data and privacy concerns or lead to misinformation and deception).

But the web of risks has many threads. "Agentic AI often functions as a ‘black box,’ which can make it difficult to explain its decisions to regulators, customers, or internal auditors," says Paul Davis, the CEO of Bank Slate, a research and consulting practice focused on banks and fintechs.

Meanwhile, automated AI decisions could inadvertently violate laws and regulations on consumer protection, anti-money laundering or fair lending laws. Agentic AI that can instruct an agent to make a trade based on bad data or assumptions could lead to financial losses and create systemic risk within the banking system. "Human oversight is still needed to oversee inputs and review the decisioning process," Davis says. "You have to monitor for AI’s blind spots in areas such as risk assessment and crisis management."

-- Article continued below --

Smaller banks face a particular challenge. Many operate on outdated technology stacks whose application programming interfaces can’t integrate with AI and that are too costly to upgrade. "Mid-size and small banks are largely playing very frenetic versions of AI Whack-a-Mole, so they’re trying out a bunch of different vendors with a specific solution to a specific problem," Reppe says. But, he added, "You have to pull yourself away from thinking there’s going to be a single vendor that solves all of my problems."

It’s an open question at this early stage whether a commercialized DeepSeek could bring down costs and level the playing field — and whether the Trump administration could deem it a threat to national security and ban its use by US companies. Meanwhile, smaller banks may find their lack of AI readiness makes them an acquisition target by larger banks seeking to grow deposits.

"There’s a lot of expectation that AI is just one of the pieces that’s going to be fueling a new wave of consolidation in the industry," Perry says.

-- Article continued below --

About the Author

Lynnley Browning is an award-winning business editor and writer who has worked at Bloomberg, The New York Times, Financial Planning magazine, and Reuters, in New York and Moscow. She has a deep background in investing, tax, personal finance, retirement, wealth management and asset management.

The Financial Brand is your premier destination for comprehensive insights in the financial services sector. With our in-depth articles, webinars, reports and research, we keep banking executives up-to-date with the latest trends, growth strategies, and technological advancements that are transforming the industry today.

© 2025 The Financial Brand. All rights reserved. The material on this site may not be reproduced, distributed, transmitted, cached or otherwise used, except with the prior written permission of The Financial Brand.