The digitally driven retail environment has caused lots of adapt-or-die challenges for payment processors and merchant acquirers, including spotting fraud as it moves from one device and channel to another.
"One of the attacks that I've seen in particular is to perform a low-value fraudulent transaction in one channel to build trust from within the system, then do a larger-value transaction elsewhere," said David Excell, founder of Featurespace, a British company that feeds artificial intelligence into behavior analysis in an attempt to track, predict and stamp out fraud that moves among channels.
"There are still friction points that should be solved," Excell said. "Just last week my credit card was blocked because of a recurring monthly insurance payment that had been on the card for four months. That's where you can apply AI. To improve spotting that and make the experience better. But there's still a long way to go."
The strategy has attracted enough attention from clients and investors to raise about $21 million in new funding in October that partly went into opening an office last week in Atlanta to handle U.S. operations.
"We have a few customers in the U.S. but we see there's a huge potential here to expand," Excell said. "Up until now we have supported these customers remotely. But we have the opportunity now to have dedicated resources on the ground."
One of Featurespace's targets is the payments processing industry—its clients include TSYS, which is rapidly entering new markets and consolidating acquired technology—and merchant acquirers. These companies have spent billions of dollars in aggregate to evolve beyond the use of point of sale hardware and on-site software deployment to a multichannel cloud-based model that accommodates mobile commerce and card not present transactions along with brick and mortar sales, marketing and payments. Featurespace's clients in Europe include Barclays and Nationwide.
Beyond mobile and e-commerce, stores are adding mobile technology that allows staff to move to help customers and take payments, or conduct outdoor sales and operate pop-up stores. That heightens the fraud risk as merchants face crime that may originate in one channel and move to another; it also stands to increase existing risks such as account takeover and ID theft. Excell contends most fraud prevention focuses on unusual transaction amounts or locations for specific users without considering the fraud may actually reside elsewhere, or the attack may be designed to circumvent behavioral analysis
"We want to move security beyond point solutions that may consider one type of transaction or one particular part of a customer's journey, and instead consider the customer's intentions, how they use a mobile device or interact with an app," Excell said.
Using artificial intelligence to feed fraud detection is an active technology play for payment companies, so Featurespace will have plenty of competitors. Socure, a New York-based startup that uses alternative data sourcing and machine learning to vet the risk of young consumers who don't have long data trails, received a funding round this past summer.
And established companies are also adding more artificial intelligence to support security.
Mastercard has taken its product, Decision Intelligence, out of pilot. This offering uses purchase value, risk profiles, location, merchant type, payment device, time and type of item to produce a risk assessment at the point of sale.
"I’m seeing a lot of growth and interest in using this type of technology for fraud detection," said Julie Conroy, a research director at Aite Group. Firms that have adopted advanced analytic platforms have achieved impressive results, she said. "Effective e-fraud prevention is an increasingly competitive issue. Early adopters of advanced analytics are able to increase their fraud detection, and the associated improvements to the customer experience give them a decided edge over their competitors that lag in these investments."
There is plenty of space for new artificial intelligence products to build a market, Excell said, contending his product uses deeper data sourcing as a differentiator.
Featurespace developed its security technology, which it calls ARIC, in conjunction with professors at Cambridge University in the U.K. ARIC monitors customer data and credit to find unusual activity. The power behind the monitoring comes from an adaptive algorithm that doesn't have to be reprogrammed manually. This technology connects to broad data sources to feed the analysis, which is designed to spot potential fraud by identifying behaviors that precede a larger ID or fraud attack, Excell said.