70% of financial institutions rely on AI and ML for fraud defence

More than two thirds of financial institutions are using artificial intelligence and machine learning to fight financial fraudsters according to a new report

The Financial Institutions Revamping Technologies to Fight Financial Crimes survey revealed that financial institutions are becoming increasingly wary of fraudsters. 40% admitted that the number of fraud incidents they’ve seen are on the rise. They’re right to be concerned. Research by UK Finance found that £1.2bn was stolen by fraudsters in 2022, the equivalent of £2,300 every minute. 

In order to combat this, many of the financial institutions surveyed for the report have adopted artificial intelligence (AI) and machine learning (ML) as part of their fraud prevention tools. The technology they’re using is mostly a mixture of in-house fraud prevention platforms, third-party resources and new technologies to protect themselves and customers. 

90% of survey respondents shared that they use fraud prevention application programming interfaces (APIs) to mitigate fraud, with 80% using adaptive and web-based multifactor authentication. 

These statistics mean booming business for financial crime prevention technology companies. When it comes to technology to alert customers to questionable activity on their account, less than half of financial institutions develop these in-house. 30% of companies develop around half of their preferred tools in-house, while just 20% develop the tools they use completely in-house. 

Perttu Nihti, chief product officer of automated invoicing technology company Basware, said: “Due to the high volume of payments, large financial institutions are prime targets for fraudsters through means such as vendor impersonation scams. An automated defence against fraud using AI and ML is essential for financial institutions, especially the chief financial officers who are ultimately accountable for any errors. 

“AI can significantly bolster the accuracy of fraud detection through sophisticated algorithms that analyse vast amounts of data to detect outliers and suspicious activity indicative of fraudulent behaviour. Not only that, but AI algorithms can be trained to minimise and reduce false positives which limits the number of legitimate transactions that are mistakenly flagged as fraudulent.” 

Image: Centropy PR

Robert Welbourn Robert Welbourn is an experienced financial writer. He has worked for a number of high street banks and trading platforms. He's also a published author and freelance writer and editor.