We might all be in the middle of one of the greatest — and most vital — cat-and-mouse games of all time.
That’s the sense one gets when considering efforts to prevent fraudsters from exploiting digital payments and commerce, efforts taking on more importance as the global economy continues to grow — and which keep getting smarter, thanks to machine learning and artificial intelligence (AI) tools.
It’s a sense one easily gets when listening to Rahul Pangam, co-founder and CEO of fraud detection services firm Simility, in a discussion with Karen Webster from PYMNTS. “Fraud detection is always a cat-and-mouse game,” he said, and that appears unlikely to change in the future, especially as ever-more emerging markets embrace digital payments and eCommerce.
PYMNTS readers likely know the stakes involved, but it never hurts to review the numbers, just so no one is in danger of becoming numb to it all. According to recent PYMNTS research, criminals last year successfully made off with $4.2 trillion from the worldwide economy, a problem that has the potential to increase as the speed of money accelerates, and more transactions move online.
Indeed, digital payments volume is increasing as emerging technologies enable more seamless online transaction experiences for consumers and businesses. Cybercriminals are also seeing opportunities to use these offerings to their advantage, employing them to exploit vulnerabilities, siphon funds and access valuable data. Those trends are sure to continue as new innovations streamline transactions.
It’s not just emerging tech that matters in this space, though — so do emerging markets, which are adopting card and digital payments, presenting fresh targets for fraudsters. One recent and significant example comes from Brazil, one of the most populous countries on the planet. In an effort to grow the usage of smartphones in Brazil, some Big Tech companies have introduced debit card payment programs. As Reuters reported, many online retailers in Brazil only accept credit cards, due to a preponderance of fraud.
There is no one-size-fits-all fraud prevention solution for all countries, Pangam told Webster. “What I would [advise] for emerging markets is very different [from] what I would build for developing markets. Brazil is very different [from] what I see in the U.S.,” he said.
In those developing markets, fraudsters can find loopholes — stemming from the fact that retailers that accept cards don’t want to put too much friction before a consumer, lest they give up. Not only that, but fraud alerts and prevention have yet to catch up to the rising tide of card and other digital payments, providing more opportunities for fraudsters.
That said, one constant does run through effective fraud prevention, which promises to gain more importance in the coming years with the rise of machine learning and AI. “Data always comes into play,” Pangam said. The trick? Using data to detect patterns that human beings cannot — even though people, thanks to what he called their ingenuity, will always have a place in the picture, a prediction that has been echoed recently by other fraud prevention technology experts.
The rising role of fraud prevention — along with the rise of digital payments in developing markets — serves to bring together technically minded and creative people to work on ways to make commerce and payments even more secure. No matter how advanced the machine learning or AI system might be, Pangam said, fraud prevention will always require human input and ingenuity.
Some fraudsters, like intelligent mice, will always find ways through anti-fraud defenses, though fraud prevention can reduce the criminals’ odds by finding those patterns ahead of time. Those patterns can be based on a variety of factors, many of them not easily discernible to human minds, Pangam noted — including, for instance, which international airports have a greater risk of fraud, or two seemingly unrelated transactions. “The data will tell you there is a connection,” he said.
Proactive moves are also key, especially in emerging markets, where merchants have not yet settled on the right level of friction to introduce into transactions, due to a fear of driving customers away.
“You have to put very basic levels of control in place,” Pangam said. That could include agents or alerts checking with the cardholder to make sure a relatively expensive or otherwise unusual transaction is legitimate. As time progresses, he noted, such checks might make use of biometric authentication, like selfie facial recognition via consumer mobile devices.
Pressure will only increase for merchants and service providers to supply even better fraud prevention, as more consumers around the world expect quicker — even instant — payments. The cat-and-mouse game continues.