Revolut has launched an advanced scam detection feature to help protect customers from card scams.

The challenger bank said since the launch of the AI-powered feature, it has observed a 30 per cent reduction in the fraud losses resulting from card scams where money has been sent for investment opportunities.

The AI-scam feature uses machine learning to detect if a customer is being scammed and can break the “spell” of the scammer before any money is sent to a criminal.

The feature, built by Revolut’s financial crime team, can decline the payment if there is a high chance that the customer is making the payment as part of a scam. The customer is then protected from making similar payments.

The customer needs to provide additional information about the transaction they are attempting to check if a scammer is guiding them through the transaction. Customers are also shown scam educational stories to urge them to think before making the transaction and can chat with a fraud specialist.

The new AI card scam feature is live in the Revolut app and available to customers worldwide.

“We’ve invested heavily in the product to ensure that customers can continue to spend and send their money safely,” said David Eborne, head of fraud at Revolut. “For example, a growing number of banks are increasingly restricting or heavily limiting the ability to make card payments to crypto and investment websites.”

He continued: “With this advanced feature, rather than completely block those transactions, we ensure that customers who want to perform legitimate payments continue to do so, but also intervene to protect those who are being guided by criminals to make fraudulent ones. We are giving our customers both freedom and security at the same time.”

Earlier this month, Revolut said the majority of scam cases reported in the UK originate on one of Meta’s platforms. The number of scams reported by its customers from Facebook, WhatsApp and Instagram grew to 66 per cent in the six months to the end of December 2023 compared to 52 per cent in the first half of the year.

---