In countless ways, digital technology has made life easier and more convenient for consumers. In at least some respects, however, it’s also made life more perilous, according to William Stern, founder of the digital small business lender Cardiff, who cites the now ubiquitous nature of cybercrime.
“It’s this never-ending sort of Whac-a-Mole of financial and identity fraud,” Stern said in an interview with The Forecast, emphasizing the growing role of artificial intelligence in financial scams and attacks.
AI fraud is on the rise:
In August 2023, six individuals were arrested in Hong Kong for using AI as part of a loan scam. According to media reports, they applied for loans online using stolen IDs and manufactured documents, including AI-altered identities, proofs of address, income statements and more.
In February 2025, Italian authorities froze nearly €1 million that cybercriminals had stolen from a businessman after assuming the identity of Italy's defense minister. Using AI to clone the politician’s voice, they had persuaded the businessman to wire funds to secure the release of Italian journalists who they claimed had been kidnapped in the Middle East.
In July 2025, scammers likewise schemed a Florida woman out of $15,000 when they used AI to impersonate her daughter in the wake of a fictitious automobile accident.
Thanks to the malicious use of generative AI by scammers, deepfake fraud has ballooned by 1,100% while synthetic identity document fraud has increased by over 300%, according to identity verification company Sumsub.
Because they house money and personally identifiable information, two things fraudsters will always covet, the surge in AI fraud is especially troubling to banks and financial institutions, suggested Sean O’Dowd, head of global financial services strategy and solutions at Nutanix.
“When people lose money, it’s not a good thing,” said O’Dowd, who has observed a growing trust gap between financial institutions and consumers as a consequence of rampant cybercrime.
“Nothing erodes trust faster than a breach of security.”
Indeed, cybercrime and AI fraud can be just as damaging to institutions’ relationships with consumers as to their balance sheets. For that reason, many financial services companies have signaled a new, broader approach to cybersecurity that encompasses not only their hardware and software, but also their trust with customers.
How FinServ Companies Can Build Digital Trust
AI tools are making it easier than ever to steal information and cause harm to financial institutions and their customers. In fact, phishing email attacks on the FinServ sector have surged 25% over the past year, AI-focused cybersecurity company Abnormal AI reported in June 2025, citing criminal use of generative AI as one likely reason.
“It’s not like the old days,” said Stern, who noted the increasing sophistication of fraudsters in the age of AI. “They’ll go through the email inbox with AI and see sentence structure, who you’re conversing with, how money is transferred and why.”
Having a digital trust strategy is one way FinServ companies can mitigate the damage caused by new and growing AI fraud threats. To develop one, industry pros like Stern recommend some core principles:
Know Your Customers
Financial services organizations should take a cue from crypto companies, which often have a “know your customer” policy, Stern said.
Historically, banks have used robust “know your customer” policies. There are 'liveness checks,' like live video to prove a customer is who they say they are. They also use continuous defense with on-chain wallet behavior analysis, which looks for changes rather than one-and-done onboarding checks.
Some users must answer out-of-wallet questions, show their ID, and take front and portrait photos. While it’s more work upfront for customers in the way of setup, it pays dividends in the form of long-term protection, noted Stern, who also advocates for employing human quality checks when appropriate. It’s similar to the “human-in-the-loop” approach being advocated in AI safety discussions.
“If paper checks are over a certain amount, it requires a second signature,” Stern said.
“If a wire or ACH is above a certain threshold, why can’t a second person give it another set of eyes? We’re in a zero-click society, but let’s get back to basics.”
Introducing intentional friction, like pulling a human into the loop, can improve fraud surveillance.
Implement Systems and Processes
Companies in quick growth phases sometimes try implementing a tool or expect employees to handle tasks without having processes in place, Stern pointed out.
That leads to confusion on both sides, with customers receiving inconsistent information that erodes trust.
“It’s not a product or vision problem. It’s their failure to systematize the business,” Stern said. “The system will always scale.”
Adopt a New Mindset
Bezalel Eithan Raviv, CEO of crypto recovery firm Lionsgate Network, said AI fraud and cyber threats will look completely different in the future. On the horizon, for example, could be new forms of embezzlement — which can be built and deployed in minutes — and constantly improving AI-generated content. IT mindsets need to change in response, he indicated.
“I don’t think you need to fight back. You need to reset,” Raviv said. “A new set of rules; new sets of security; and new sets of how you store, identify and authenticate.”
Use Tools and Software Strategically
Raviv said many companies are spending time and money on outdated or irrelevant security methods without considering how it might impact their human customers.
Though technology is essential to closing the digital trust gap, IT leaders must approach any tools with a customer-first mentality, he stressed.
“They haven’t considered new threats; they’ve considered cloud threats. Cloud protecting cloud. Does it protect people?” Raviv asked.
“You can say you have the most advanced things, but without that, it’s like having the best door while the wall on the other side of the room is wide open.”
Communicate with Respect
AI fraud and financial scams are getting more complex, yet customers also are more savvy. Transparency and openness go far in good times and bad, according to O’Dowd.
“The business of banking may be to make money, but banks need their customers to trust them enough to put money in their care. Then, it’s the bank’s responsibility to make sure customers’ money is available to them when they need it.”
Consider the May 2025 cyberattack on the crypto platform Coinbase, which cost the firm up to $400 million in compromised customer information. In that situation, the company offered an open look at what happened. It transparently shared what the extortionists got, what they didn’t and how the company was handling it, including reimbursing customers who had been tricked and advising on improved protection strategies. That clear playbook, featuring a “tl;dr” (too long; didn’t read) overview, kept customers in the know and helped re-establish trust after a rocky situation.
Pairing Humans and Technology in Financial Services
Financial services are leaning heavily into AI, with technology investments and roadmaps expanding. However, not everything needs to involve generative AI, noted O’Dowd, who highlighted how transactions still sit on mainframes because they’re cheap, consistent and effective.
The sweet spot is finding where generative AI improves or allows for use cases that were previously harder or impossible to execute. For example, natural language processing has advanced so much that banks can leverage generative AI for research and text-based use cases. Machine learning, meanwhile, remains a great tool for more traditional, quantifiable use cases, including fraud detection, credit scoring and algorithmic trading.
O’Dowd underscored that while enthusiasm for new technology is healthy, companies don’t need to overshoot when developing objectives.
“When I was leaving the capital markets industry less than a decade ago, fax was still used in the departments,” he said. “You don’t need the height, you just need to apply what is realistically within reach.”
Beyond digital identity verification, FinServ companies are implementing secure recording of audio-visual sessions and offering tamper-evident document protection. These features can help keep sensitive data safe while alerting companies to potential threats before they become too damaging.
Other tools offer relief for business units, giving employees more time to focus on strategic tasks and critical thinking. At Lionsgate, for example, a goal is providing navigation to help companies wade through potential threats.
“We’re working on prevention instruments to read with you and give you the guidelines, just like Waze,” Raviv said.
“You know the way, but you don’t want to think about it, so you can just follow a path.”
Even with an influx of AI tools and software, humans are essential in preventing fraudulent actions, according to Anant Agrawal, president of the online notary platform OneNotary. The act of notarization isn’t just to verify an individual’s identity, for example. It’s also to confirm that an individual knows what they’re doing and whether they’re doing it under duress.
“An online notary or human in the loop is really valuable because it’s easy. It’s quick to put in there and doesn’t add a lot of friction,” Agrawal said.
“You can lie to a human being and not go to jail. You lie to a notary, a government-certified official, it’s a felony, and you can go to court. That fear factor plays a role in limiting bad actors or people trying to game the system.”
Ultimately, Agrawal suggested, human teams working together with AI tools will offer the greatest form of success and help companies maintain a high level of digital trust, even as cyber criminals, scammers and AI fraud undermine consumer trust more broadly.
“Technically, notaries are fraud fighters. We’re giving them AI tools to help them do their jobs a lot better,” Agrawal said of the online notaries that use his platform. “We’ve turned them into RoboCops.”
Joey Held is a writer, podcaster and author based in Austin, Texas, and the founder of Fun Fact Friyay and Wacky Travel. Connect with him on X or LinkedIn.
© 2026 Nutanix, Inc. All rights reserved. For additional information and important legal disclaimers, please go here.
Related Articles
