- Mastercard, an international payment services company, works with banks all over the world.
- The company is using AI to try to identify when customers are likely to fall victim to fraud.
- This article is “CXO AI Playbook“Business leaders speak openly about how they're testing and using AI.”
In “CXO AI Playbook,” Business Insider presents mini-case studies of AI adoption across industries, company sizes, and technology DNA. Each company talks about the problem they're trying to solve with AI, who is making those decisions within their company, and their vision for the future of AI.
Mastercard has been providing payment card services to the world's largest banks and millions of consumers since 1966. The company responds to the ever-evolving threats posed to its customers by fraudsters and criminals.
Situation analysis: What problem was the company trying to solve?
“For decades, payments fraud has involved using card information purchased, found or counterfeited in person or online,” Chris Reed, executive vice president of identity solutions at Mastercard, told Business Insider.
But payment fraud trends are changing, and banks and card companies need to respond. Reid says one major new trend in fraud is what's known as authorization push payment fraud, where a fraudster poses as a trusted person, such as a bank, official authority or loved one, to convince someone to send money.
It's hard to tell when a cardholder is about to send a large sum of money to a fraudster and then lose it, Reid said. “How do you know that you're a victim of a crime, or about to become a victim?” he asked. The company believes artificial intelligence can help it identify when fraud is likely to occur.
Key Staff and Partners
Development of the AI fraud detection system began as a pilot program in the U.K. At Mastercard, teams working on the project included Reid's identity group, the security and cyber innovation team, and the open banking team, which is working on connecting competing banks.
Reid said the job could not have been done without UK banks including Lloyds, NatWest and TSB putting aside rivalries to work together.
“This only works if we all work together,” Reid said. “The good thing we see when it comes to fraud and security is that banks are competing to win every day. When it comes to fraud and security, we're competing to beat the fraudsters.”
Use of AI
With the support of these banks, the team at Mastercard began developing an AI system that would identify signs that someone was about to be scammed on both the victim’s and the fraudster’s accounts.
“We are providing bank customers with a three-dimensional view of the two accounts at issue,” Reid said.
Mastercard's AI systems are trained on data from both legitimate and fraudulent payment activity and use this information to monitor and identify suspicious transaction behavior, examining data such as payee names to identify suspicious transactions.
You can also look into any second and third degree relationships linked to the fraudster's bank accounts.
“Criminals need to find a way out of the legitimate financial ecosystem,” Reid said. “Money tends to flow into various mule and drop accounts.”
The system provides real-time AI-powered risk scores for transactions and accounts, alerting banks if Mastercard suspects a payment is going to a known fraudster.
Did it work, and how did the leaders know?
Reid said the TSB had been “very transparent in helping us understand the effectiveness of the solution.”
Based on the scale of fraud prevented among TSB customers and extrapolating that to the whole UK banking industry, “all other things being equal”, the AI system Reid told BI that losses in the UK alone were an estimated 100 million pounds ($126 million).
He added that Mastercard's technology is “very good” at identifying “identity fraud, romance fraud, purchasing fraud.”
What's next?
Mastercard needs to take a stand against fraudsters, Reed said.
“Fraud is an incredibly successful industry,” he added. “Fraudsters invest in product development and in making themselves more efficient.”
That's why Mastercard plans to expand its AI system, which currently provides AI-identified risk scores for potential victim accounts. By the end of 2024, Mastercard hopes to provide risk scores for suspected fraudsters to raise awareness for bank customers.
“This doubles our ability to stop funds from getting into the hands of criminals,” he said. “We can tell the fraudster's bank, 'This is a high-risk transaction. You might want to delay withdrawing the funds.'”
Mastercard is also working to develop more practical applications for the risk scores, which could be useful to senior bank executives but not to rank-and-file bankers tasked with helping victims.
“In this case, contact center agents are trained and knowledgeable about fraud, but they need to be able to turn the scores into something useful,” he said. “We're working to ultimately stop this crime.”
We'd love to hear from you, so if you have any thoughts about your company's AI journey, please email us. jhood@businessinsider.com.
