How Financial Crime Is Being Thwarted By AI, ML

Despite the fact that they prevent over $22 billion in cybertheft annually, financial institutions (FIs) are seeing fraud attack volumes soar as cyberthieves look for new ways to invade. 

From account takeovers (ATOs) — which have been staggeringly popular with crooks during the pandemic — to brute-force attacks, it’s a cyber jungle out there … and companies need protection. 

“FIs are turning to advanced technology to arm themselves against fraudsters’ seemingly endless supply of stolen data,” according to PYMNTS’ new Preventing Financial Crimes Playbook done in collaboration with NICE Actimize.

“Some of the most promising solutions come in the form of artificial intelligence (AI) and machine learning (ML), which can analyze thousands of ongoing transactions and applications every second and pinpoint telltale signs of fraud, like uncharacteristically large transactions or login attempts from multiple devices in different regions.”

The new Playbook states that “banks have reported increased fraud detection rates of up to 50 percent through the use of these technologies, but the tools are still used by only a minority of FIs around the world.” 

Increasing the adoption of crime-fighting solutions in a world of “digital everything” is the next step.

A ‘Holistic Response’ to Customer Security

With phishing scams now more convincing than ever, leaders in the cybersecurity field believe this is a time to redefine online crime, and to adopt new tools and technology to defeat it.

“We are starting to see the need for fusion between ‘cyber’ and ‘fraud.’ By bringing elements of each discipline together as part of a holistic response to customer security and fraud prevention, we can see important improvements in detection and prevention,” Yuval Marco, general manager of fraud and authentication at NICE Actimize, told PYMNTS.

He noted that this starts with company-wide password policy review, along with the integration of authentication and fraud profiling, to score enrollments and logins as well as payments.

“By feeding in all the relevant events, normal customer patterns and profiles can be built. This allows for more friction weighted to the risk of the event at hand, and MFA [multifactor authentication] can be directly linked to the level of risk, increasing security while allowing genuine customers through,” Marco said. “Once that’s complete, enrich your fraud profiling system with data and intelligence, including credentials from internal and external sources.” 

This approach is often lethal to bad attack types like credential stuffing, but actually lessens frictions for known entities and individuals. That’s how FinServ friction should work.

Superpowers for Financial Crime Fighters

Teams of humans have been ferreting out financial crime for decades. Now, a bit like the Agents of S.H.I.E.L.D., they have superpowers to help them in the fight. 

“Banks are deploying AI-based systems in record numbers, with more than $217 billion spent on AI’s applications for middle-office use cases like fraud prevention and risk assessment. These investments are paying off,” according to the new Preventing Financial Crimes Playbook.

Some 80 percent of experts report that AI reduces payments fraud and almost 64 percent of FIs say AI is vital to stopping fraud in action. “These systems are commonplace at large FIs that have more than $100 billion in assets — 72.7 percent of which leverage AI — but only 5.5 percent of all FIs reportedly have an AI-based system in place.” 

That leaves plenty of room for improvement. The good news is that FIs are making these investments as the post-pandemic business world slowly reorganizes around speed and safety.