Visa is using AI tools to ramp up its fight against fraud, as financial institutions face mounting threats in an increasingly digitalized payments environment. 

Instances of document forgery, for example, increased 244% year over year in June as fraudsters use AI to alter documents like loan applications and government-issued ID’s, according to Visa’s recently released report, “Fight fraud with leading-edge strategies to stay ahead of complex threats.”

Identification documents are the most common type forged, the report found.  

Advanced technology, like generative AI, makes it easier for fraudsters to forge documents, and this is reason enough for the industry to shift from physical documentation, Kate Manfred, North America head of advisory services at Visa, told Bank Automation News. 

“While there is no silver bullet in the fight against fraud, adopting a multi-layered defense strategy is key to combatting the various forms of fraud,” Manfred said. 

Early warning signals 

To combat fraud, Visa’s clients are deploying the following AI technologies from the payments giant: 

• Device intelligence; 

• Geolocation inconsistencies detection;  

• Behavioral analytics; 

• Digital ID proofing; 

• Machine learning-driven risk scoring; and 

• Use of fraud databases to cross-reference applicant information. 

AI-driven, automated and integrated technology platforms can effectively flag risks and route alerts to appropriate teams, ensuring timely and coordinated responses, Omar Akkor, senior director of banking market strategy and innovation at Moody’s, told BAN. 

In fact, many banks have established common early warning signs to identify potential risks when multiple contradictory signals are triggered simultaneously, he said  

“These AI systems not only detect warning signals but also interpret them, providing context and suggesting potential actions when specific thresholds are breached,” Akkor said. 

Register here for “Seamless integration: The new frontier in embedded payments,” a free Bank Automation News webinar set for Tuesday, July 22, at 11 a.m. ET.