AI is transforming fraud prevention in digital payments. Here’s how.
With the growth of e-commerce transactions and digital payments, the world is encountering new challenges in thwarting fraud. Research predicts that merchants worldwide will lose more than $362 billion to online payment fraud between 2023 and 2028.
In the face of this staggering financial loss, AI-driven tools are changing the landscape of robust fraud prevention. These risk solutions harness the power of artificial intelligence to help safeguard merchants, financial institutions, and consumers at every transaction stage.
New tech means more sophisticated fraud
James Mirfin is the SVP and global head of risk and identity solutions at Visa. He shares that AI and other cutting-edge tech offer fraudsters brand-new tools, plus faster methods to employ tried-and-true techniques.
Mirfin noted that AI and bots have made card enumeration, or brute force attacks, vastly more efficient — amounting to more than $1 billion of annual fraud losses. Criminals’ bots can generate massive lists of potential credit card numbers, then leverage AI to rapidly test these values in online transactions until it finds a valid set of card data.
Beyond duping payment systems, AI has become increasingly effective at scamming real people. “Fraudsters are using generative AI (Gen AI) to conduct phishing attacks or socially engineer people into making payments or transferring funds to a fraudsters’ account,” Mirfin said.
The impact of fraud is sobering. Consumers can face devastating monetary damage, often coupled with the shame of being deceived. “For merchants and financial institutions,” Mirfin said, “it can result in financial losses, a significant operational burden, and the need to invest in people, processes, and technology to protect themselves and their customers.”
Modern challenges for financial institutions
The rise of e-commerce means an increase in card-not-present (CNP) transactions. CNP transactions provide unique challenges, as they lack the security of physical chip cards, cardholder ID, and chip cryptography.
Strong security measures can limit fraudulent transactions. However, such measures run the risk of being overzealous by refusing valid transactions. Studies show that 31% of cardholders are more likely to end a banking relationship after experiencing fraud. But 51% of customers switch cards mid-transaction if their first attempt declines.
“These stats,” Mirfin said, “highlight how difficult it is for financial institutions to find the balance and precision needed to stop bad transactions without incorrectly impacting a good customer when they do a genuine transaction.”
AI shines in fraud prevention
For over 30 years, Visa has been a leader in using AI to combat fraud. Today, Visa probes billions of transactions each year. Born of that powerful analysis, the comprehensive Visa Protect suite of solutions empowers merchants and financial institutions — helping them prevent e-commerce fraud and grow their businesses with seamless, protected transactions.
Mirfin shared that Visa Protect offers rapid, real-time risk scoring at every stage of a new transaction. Visa’s risk scores assess the likelihood of fraud in token provisioning events, authentication requests, authorization requests, card-not-present authorizations, and credential enumeration.
But Visa leaves the decision-making power in the hands of merchants and banks. “Financial institution clients,” Mirfin said, “can take the risk score at each step of the transaction journey, assess their own risk tolerance, and determine whether they want to approve or decline each step of the transaction to mitigate risk throughout the transaction journey.”
Mitigating payment risk in the modern age
In fiscal year 2023, Visa prevented $40 billion in fraudulent transactions. Its current suite of value-added services comprises almost 200 products across the sectors of Acceptance, Advisory, Issuing, Open Banking, and Risk and Identity.
Current* AI products in Visa Protect include the following:
- Cybersource Decision Manager for merchants
- Visa Provisioning Intelligence to protect financial institutions’ token provisioning requests
- Visa Consumer Authentication Service for issuers’ authentication strategy
- Visa Deep Authorization for CNP payments for issuers
- Visa Advanced Authorization and Visa Risk Manager for financial institutions’ Visa and non-Visa card payments
- Visa Protect for Account-to-Account Payments for financial institutions
- Visa Account Attack Intelligence Score to protect issuers from enumeration attacks
Visa’s risk-scoring methodology continues to evolve. To meet the challenge of constant advances in fraud tactics and technology, these models are routinely evaluated and improved with relevant data — meeting the challenge of constant advances in fraud tactics and technology.
“Payments have changed more in the past five years than in the last 50,” Mirfin said. “We’ve barely just scratched the surface with regards to what AI can do as it relates to eradicating fraud and creating a more seamless, secure payment experience. Visa is keeping pace by bolstering our powerful suite of Visa Protect risk solutions through unmatched technology, responsible AI expertise, and expanding utility beyond the Visa network.”
Protecting your business and your customers
Merchants and financial institutions face a growing threat from new fraud techniques and technology. But AI has proven to be a powerful weapon in risk prevention for digital payments and transactions. Modern businesses can rely on Visa Protect’s comprehensive solutions to safeguard their online data and ensure smooth customer interactions.
Looking for comprehensive payment protection? Learn more about Visa Protect risk solutions.
