AI In FinTech: How to Build Trust, Not Just Efficiency

Financial institutions have poured billions into artificial intelligence over the past two years, and the payoff is starting to show. Fraud detection is sharper, customer service is faster and personalization has moved well past generic marketing emails. Yet those gains have exposed a deeper challenge: an IDC study commissioned by Atlanta-based Sage found that 71% of finance leaders would reject an AI tool that was 99% accurate if it couldn’t explain its answers. More than half would pay a premium for AI that shows its work. For every leap forward, executives are discovering how much further the industry has to go when it comes to transparency, customer trust and understanding people rather than simply transaction histories.

To make sense of where AI is truly moving the needle in financial services, and where the industry keeps stumbling, we turned to members of the Senior Executive FinTech Think Tank, a curated group of financial technology leaders and advisors. Their answers point to a pattern: the easy wins are technical, and the harder, more valuable work is human.

Here, they share where they see AI creating meaningful change today, and where they believe the industry is still getting it wrong.

“The industry still needs to do a lot better with AI in transparency, especially with offboarding, fine print and legalese, which still hit consumers when it matters the most.”

– Banu Raghuraman, AI Lead Product Manager at Perficient

SHARE IT

Fraud Detection Has Become AI’s Clearest Win

Few corners of financial services have benefited from AI as visibly as fraud prevention, according to Banu Raghuraman, AI Lead Product Manager at Perficient, a global digital consultancy helping banks and other financial institutions build AI-powered products and customer experiences. Raghuraman has spent more than 15 years building customer experiences and software solutions across fintech, payments and healthcare, giving her a front-row view of how AI performs once it reaches real customers.

“Few of my favorite use cases are in fraud, personalized account oversight and personal financial management,” Raghuraman says. Fraud, she notes, has always relied on analytics, but AI has taken detection to a new level. For instance, Mastercard’s proprietary model, released in 2024, was designed to improve fraud detection by up to 300%, according to the company’s reporting.

Account monitoring has come a long way, too, even if the underlying idea isn’t new. “Capital One released an AI assistant, Eno, back in 2017, well before the current GenAI wave, but the quality of user interactions and alerts has greatly improved and makes you feel like you have an assistant overseeing your accounts,” Raghuraman says. She also points to personal financial management platforms as examples of how AI has transformed a familiar category, layering smart spending insights, personalized recommendations and contextual financial guidance onto tools that once offered little more than static budgets.

But Raghuraman is clear that the progress has limits. “The industry still needs to do a lot better with AI in transparency, especially with offboarding, fine print and legalese, which still hit consumers when it matters the most,” she says. In other words, the moments when customers are most vulnerable, such as closing an account or parsing a disclosure, are often the moments AI has done the least to clarify.

“For decades, financial institutions have built products around demographics, balances and transactions, yet they still know very little about the human being behind the money.”

– Tamara Kostova, Founder and CEO of AllVesta

SHARE IT

Making Finance More Human Is the Bigger Opportunity

For Tamara Kostova, Founder and CEO of AllVesta, the operational wins from AI are real, but they aren’t the whole story. AllVesta is building the behavioral intelligence layer for retail investing, helping banks, asset managers and wealth platforms understand the person behind the money so they can turn savers into confident, long-term investors. Kostova brings more than 20 years of experience across capital markets, banking and financial technology, including senior roles at DXC Technology, Deutsche Bank, UBS and Thomson Reuters, and she previously founded and scaled Velexa, which is a WealthTech 100 company that expanded access to digital investing infrastructure before its 2025 acquisition.

“AI is already creating significant value across financial services by automating operations, improving compliance, accelerating software development and increasing organizational productivity,” Kostova says. “Those gains are real, but I believe the industry’s biggest opportunity lies elsewhere: making finance more human.”

That opportunity, she explains, comes from a gap most institutions have never closed. “For decades, financial institutions have built products around demographics, balances and transactions, yet they still know very little about the human being behind the money,” Kostova says. Financial decisions, she points out, are shaped by confidence, emotion and behavioral bias as much as by raw information, and for the first time, AI gives the industry a way to understand those factors at scale and deliver personalized financial experiences.

That is precisely where Kostova believes many institutions are misapplying the technology. “Where I think the industry is getting it wrong is treating AI primarily as a tool for internal efficiency and cost reduction,” she says. “The real transformation will come when institutions use AI to improve customer outcomes, helping people build confidence, make better financial decisions and ultimately create more wealth.”

Where Leaders Should Focus Next

Taken together, these insights suggest that AI’s clearest returns come from operational efficiency, while its greatest long-term opportunity lies in strengthening customer relationships.

• Treat fraud detection as a foundation, not a finish line. AI has already made real-time fraud monitoring dramatically more accurate. That same rigor must extend to the plain-language moments, like account closures and disclosures, that customers actually read.
• Use AI to understand people, not just their transactions. The institutions that go beyond efficiency gains and apply AI to customer confidence, financial literacy and long-term outcomes are the ones most likely to convert access into genuine engagement.

The Human Advantage Will Define AI’s Next Chapter

AI has already proven it can outperform static rules and manual review in areas like fraud detection, and it has made everyday account monitoring feel more like having a genuine financial partner. Those gains are worth recognizing, but they are also the easiest problems for AI to solve, because they involve patterns, not people.

The harder and more valuable work, as both Raghuraman and Kostova suggest, lies in using AI to close the gap between what institutions know about their customers and what actually drives their financial decisions. The organizations that treat AI as a tool for building trust and understanding, not just efficiency, will be the ones that turn today’s technical wins into lasting customer relationships.