From scam phone calls, text messages and emails, fake websites, falsified documents and fake insurance claims, fraud has become big business – and a big problem.
In the first half of 2021, criminals stole a total more than £753m million through fraud, an increase of 30 percent compared to the first half of 2020, according to data from banking industry body UK Finance.
The problem is only going to get worse, according to Martin Rehak, founder and CEO of Resistant AI, a firm whose artificial intelligence technology helps detect malicious behaviour.
“Today’s fraud problem is primarily one of scale. Traditional methods of tackling the problem that rely largely on human intervention are like taking a knife into a gun fight”, Rehak told City A.M.
“Tackling fraud is becoming progressively more complex, thanks to the sheer speed at which fraudsters are adapting and refining their methods.”Martin Rehak
This is putting traditional banks and fintechs alike on the back-foot when it comes to proactively spotting fraud patterns within the huge volume of transactions that they process daily.
Paypal’s recent disclosure that it had 4.5m illegitimate accounts on its network is a perfect example of the challenges companies face.
“Fraud and cyber analyst teams are simply overwhelmed by the ever-increasing number of alerts, many of them false,” Rehak pointed out.
AI and machine learning
With a PhD in Artificial Intelligence and more than two decades developing intelligence-based security solutions, Rehak knows first-hand that to effectively tackle financial crime, the only solution lies in the use of sophisticated artificial intelligence and machine learning.
“Sophisticated AI is able to predict, detect and defer financial crime by using multiple algorithms to detect patterns of fraudulent behaviour in real time.”
Today’s sophisticated AI technologies are great at spotting anomalies, Rehak continued.
These could be behaviours that deviate from the norm or relate to device or contact information changes and unusual switching between accounts. They are all indicators of criminal activity, he explained.
“The combination of AI, automation and the human brain is the strongest form of defence against cybercrime.”Martin Rehak
“When an alert is triggered, experienced human analysts can then step in to infer the attackers motives and react appropriately.”
“At the same time, the vast majority of valid transactions are allowed to proceed as the AI techniques are very precise in separating good from bad.”
The finance sector will need to work harder to mitigate risks as fraudsters shift methods and change targets, Rehak argued.
“The pace at which the threat landscape is evolving means companies must adjust their approach if they’re to successfully crack down on fraud.”
Rehak is convinced that fraud is here to stay and criminals are getting smarter.
“They are getting hold of our identities to make purchases, they are taking over our bank accounts, they are making fraudulent insurance claims in our names and they are scamming us,” he concluded.