Using AI to stay ahead of payment fraud
Fraudsters are launching faster, more sophisticated scams. Visa Protect combines AI-driven tools with network intelligence to flag suspicious activity while keeping genuine payments moving.
by Brand Studio · CNA · JoinRead a summary of this article on FAST.
Get bite-sized news via a new
cards interface. Give it a try.
Click here to return to FAST Tap here to return to FAST
FAST
When consumers tap a card, shop online or transfer money, they expect the transaction to be secure.
As more payments move online, maintaining that trust is increasingly important for businesses. A Visa-commissioned study found that close to seven in 10 Singapore residents trusted digital payments, while 42 per cent had encountered a scam and 8 per cent had lost money to one.
Fraud tactics are changing, too. Rather than only intercepting transactions, scammers are increasingly persuading victims to authorise payments themselves, using artificial intelligence (AI), automation and social engineering.
This puts businesses in a bind. The same AI that helps them personalise shopping recommendations and speed up payment processing can also be used by fraudsters. For banks, merchants, fintech companies and other payment providers, the challenge is to spot suspicious activity quickly without slowing legitimate transactions.
04:01 Min
Mr Varun Mahindru (left), Visa Asia Pacific’s fintech business development lead, and Mr Phillip Finnegan, Featurespace’s general manager for Asia Pacific, discuss AI fraud detection. (Video: Visa)
WHY TRADITIONAL FRAUD CHECKS ARE FALLING BEHIND
Fraud prevention has long relied on fixed rules that flag or block risky transactions. These checks remain important but can struggle to keep pace with new fraud tactics and digital channels. Increasingly, criminals are not just intercepting payments but impersonating the people making them.
“Fraudulent activities and scams are moving from transaction to identity,” said Ms Adeline Kim, Visa’s group country manager for regional Southeast Asia and senior vice president for global clients in Asia Pacific. “Scammers can now tailor exactly how and when they strike to make it convincing.”
Undetected fraud can cost businesses money, disrupt operations and damage reputations. But overly strict checks can block legitimate transactions – frustrating customers and potentially sending them elsewhere.
Ms Kim said fraudsters were also using technology to target weak points more quickly.
“For banks, merchants and fintechs, the concern is no longer just how much fraud there is, but the speed and scale at which it can strike,” she added. “The response has to be just as fast, intelligent and coordinated.”
DEFENDING AGAINST FASTER THREATS
Visa has invested more than US$13 billion (S$16.8 billion) in technology over the past five years, including fraud prevention. It has used AI to fight fraud for more than three decades, analysing more than 250 billion transactions each year based on about 500 data points per transaction.
Visa Protect* brings these capabilities together in a portfolio of AI-driven fraud prevention solutions for issuers, merchants, acquirers and fintech companies. Its tools assess transactions in real time and compare them with patterns across Visa’s global network, helping businesses decide how to respond to different warning signs.
The portfolio includes technology from Featurespace, which Visa acquired in 2024. The technology learns each customer’s typical behaviour and flags activity outside those patterns. The models continue learning over time, helping identify fraud techniques that traditional rules may miss.
“Hundreds of rules are costly and ineffective in the current environment,” said Mr Phillip Finnegan, Featurespace’s general manager for Asia Pacific. He added that businesses should move towards machine learning and AI-based fraud capabilities.
Visa recently updated its Singapore Security Roadmap, first launched in 2019, with priorities including stronger authentication, tokenisation and closer industry collaboration against fraud and scams involving AI.
SAFER PAYMENTS WITHOUT ADDED FRICTION
Fraud checks need to catch suspicious activity without blocking genuine payments. Yet legitimate transactions can still be declined when customers are travelling or making important purchases, even as businesses need stronger controls against increasingly sophisticated scams.
Traditional systems may flag anything unusual as suspicious. AI-based systems can assess a transaction in a broader context, including a customer’s spending patterns and activity across the payments network. This can help distinguish genuine purchases from fraud attempts, reduce false declines and allow fraud teams to focus on higher-risk cases.
“The goal is never to take customers out of the payment process. It is to remove unnecessary friction while keeping them protected,” said Ms Kim.
As digital commerce expands, maintaining that balance becomes increasingly important.
“Payments are ultimately about trust between buyers and sellers who may never meet,” she added. “Getting security and trust right is not only about stopping fraud. It is about giving customers the confidence to keep transacting.”
Learn how Visa Protect helps businesses across the payments ecosystem manage fraud risk in real time.
* Visa Protect includes Visa Advanced Authorisation for real-time transaction risk assessment, Visa Risk Manager for issuer fraud controls, Decision Manager for AI-based fraud detection in digital commerce and Featurespace for behavioural analytics that flags suspicious activity.
Sign up for our newsletters
Get our pick of top stories and thought-provoking articles in your inbox
Get the CNA app
Stay updated with notifications for breaking news and our best stories
Get WhatsApp alerts
Join our channel for the top reads for the day on your preferred chat app