As banking has become increasingly digital, so too has financial crime – unauthorised financial fraud losses across payment cards, remote banking and cheques totalled £783.8 million in 2020 in the UK alone.
Fortunately, as cybercrime has evolved, fraud prevention techniques have too. There are now a number of sophisticated, AI-powered solutions that make it easier for banks to protect customers, avoid fines and safeguard their reputation. Let’s look at 3 trends in this area.
The crime: Wire fraud
80% of adults in the UK use some form of digital banking, and 27% have made the switch to a fully digital bank. This makes it easier for scammers to extract money quickly.
The solution: Transaction vetting and automated alerts
Given the sheer volume of daily digital transactions, banks can’t manually vet every transfer. Instead, they can use AI-powered automated systems to identify instances of possible fraud.
These systems continuously scan data passing through and use pre-programmed alerts and machine learning to flag suspicious activity. Bank employees can then examine these suspicious transactions to reduce the risk of blocking legitimate transactions. Customers can also receive automatic fraud and suspicious activity alerts, which customers can respond to in real-time to verify and authorise legitimate transactions or conversely report suspected fraud.
The crime: Credential stealing
In the days of physical banking, it was much harder to impersonate someone. Now, in the digital arena, it’s easier than ever to pose as someone else.
Fraudsters use increasingly sophisticated phishing techniques to trick people into parting with personal data, including card information, passwords and answers to security questions.
The solution: Biometric authentication
Strong passwords and heightened awareness can help keep customer accounts safe, but even the savviest consumers can fall victim to phishing.
To add an extra layer of security to identity and verification (ID&V) processes, banks can use multi-factor authentication. Sending a one-time password (OTP) to a user’s registered device is a common approach here, but AI-powered biometric authentication is arguably more robust, because it’s much harder to defraud. Examples include fingerprint and facial recognition, as well as voice recognition that detects the unique cadence of a person’s voice.
The crime: Account takeover
When a person’s security credentials are stolen, the best-case scenario is that their password is reset. The worst-case scenario is account takeover and identity theft.
Unfortunately, this type of crime has become more common since the pandemic because more people have moved to using online services.
The solution: Machine learning using consortium data
Intelligent systems use machine learning to help spot fraudulent activity, which means they continuously improve as the quantity and quality of the data it analyses increases. A valuable way to accelerate this improvement is to use consortium data – which accounts for data pulled from multiple sources, including other banks.
When financial institutions collaborate to share fraud data, they create a database of known threats. This gives AI-driven systems more examples of potential fraud. When they think an account or transaction might be compromised, they look for specific behavioural cues based on the in-depth data trends. They then automatically alert customers and account managers, who can look more closely and act if necessary.
Financial losses due to fraud can cost banks dearly – not only in money, but in productivity, reputation and customer relationships. By investing in AI-driven technologies, you can protect customers in an increasingly digital-first world.
To find out more about AI-driven fraud management, download our recent whitepaper.