Fraud Detection Systems Explained

When it comes to digital financial services, security must be top of mind. That’s why as the Central Bank of Barbados develops its national instant payment system, BimPay, it has incorporated many features to keep users’ data – and hard-earned money – safe. 

Among these features are multifactor authentication, encryption, and AI-powered fraud detection. To ensure that you’re familiar with these tools and how they work, we contacted Anthony Harris, President of the Barbados Chapter of the Information Systems Security Association, and asked him to walk us through them for us.

In previous articles, he explained multifactor authentication and encryption. This time around, he’s breaking down fraud detection.

What is fraud detection in the context of an online financial system?

Fraud detection in an online financial system means using tools and technology to spot unusual or suspicious activity, like someone trying to steal money or use your account without permission, so the system can stop it or warn you before any damage is done.

How is AI used to assist in fraud detection?

AI helps with fraud detection by quickly analysing huge amounts of transactions and spotting patterns that look suspicious or out of the ordinary. It can learn what normal activity looks like for each person and flag anything unusual, helping banks and companies catch fraud faster and more accurately than humans alone.

What are some of the things fraud detection systems look for?

Fraud detection systems look for things like unusual spending habits such as a larger than typical purchase, multiple failed login attempts, or very quick, successive purchases that don't fit your normal pattern. They're basically looking for anything that seems "off."

Are there legitimate activities I might engage in that might get flagged as potentially fraudulent?

Yes, definitely. For example, if you make a big purchase you don’t usually make or shop online from an unfamiliar website, these can sometimes get flagged as suspicious even though they’re totally legitimate. The system just wants to be extra careful.

Special thanks to Anthony Harris and the entire team at the Barbados Chapter of the Information Systems Security Association.