How machine learning can stop terrorists from money laundering

ISIS, among others, are using complex schemes to clean up funds - and it's up to us to find solutions to the problem.
Written by Charlie Osborne, Contributing Writer
Stock photo

Big Data and analytics will become a vital tool in detecting and preventing advanced money laundering schemes used to fund terrorist activity.

For criminal operations to run and for terrorist organizations to receive the funding they need to operate, money is required. We've come far from the days of cloak-and-dagger meetings in person and dodgy deals in the shadows when it comes to money laundering; and instead, law enforcement and banks are facing complex schemes which are becoming increasingly difficult for humans to both detect and prevent.

In an interview with ZDNet, Mark Gazit, CEO of cybersecurity firm ThetaRay said that while billions of dollars are lost through cyberfraud every year, we can "safely assume" that a "significant" portion of this amount involves money laundering.

It is not as difficult as you may think to launder money online. For example, a quick look through the Dark Web and you can find countless "washers" which, for a small fee, will take user Bitcoin and "wash" them clean, funneling them so the origin of these funds cannot be traced.

The executive says that terrorist groups, such as ISIS, are increasingly using these kinds of methods in fresh money laundering schemes.

One new method, for example, is the use of cyberattacks in which operators steal no more than $1 from a bank account - but this occurs automatically to millions of accounts at a time. As it is such a small amount, it remains unnoticed by banks or account holders themselves, and this can lead to millions of dollars being transferred to terrorist organizations.

"These groups know that they can't just sell oil and receive a lump sum of 10 million dollars; it would be identified as money laundering and intercepted by law enforcement agencies," Gazit says. "But through a flood of micro transactions, they can escape detection, acquire the necessary funds, and continue their terror operations."

Other methods of money laundering include capitalizing on online gaming and virtual currencies, "carding" - the transfer of money to card sellers - and the use of money mules to withdraw funds.

10 things you didn't know about the Dark Web

The Internet and the evolution of computers and networks have proven to be a catalyst for economic growth, jobs, and advances in everything from research to education worldwide.

However, it has also led to an expansion in cybercrime. On a global platform, it can be a difficult task for law enforcement to not only track down the origin of criminal activity - and any money laundering schemes connected to it - but also those ultimately responsible.

Gazit commented:

"A single person can now use technology to break into millions of machines and conduct money laundering activities. Technology not only allows it to happen automatically; it significantly lowers the risk faced by the hacker. He can be offshore in another country, using the Internet to break into bank accounts around the world.

Even in the worst case scenario, all that happens is that the bank detects the activity and stops the transfer.There is very little risk of him being captured and jailed."

If you can't necessarily track down the origin of such activity, the next best thing is to be able to detect it quickly and shut it down before networks are compromised or financial losses occur.

According to the executive, terrorism-related money laundering schemes can be stopped through the employment of smart machines, Big Data, and analytics.

"If terrorists are going to use smart machines to conduct automated crime, it's in our best interest to employ those same machines in our fight to stop them," Gazit says. "Humans are simply not equipped to detect and prevent this sort of fraudulent activity."

By taking advantage of Big Data, machine learning systems can process and analyze vast streams of information in a fraction of the time it would take human operators. When you have millions of financial transactions taking place ever day, ML provides a means for automated pattern detection and potentially a higher chance of discovering suspicious activity and blocking it quickly.

Gazit believes that through 2017 and beyond, we will begin to rely more on information and analytics technologies which utilize machine learning to monitor transactions and report crime in real time, which is increasingly important if criminals are going to earn less from fraud, and terrorism groups may also feel the pinch as ML cracks down on money laundering.

The executive says that we will definitely see an increase in the use of technology for criminal gain in the future, whether this be terrorism or the use of ransomware to blackmail both businesses and consumers out of their funds.

However, if we use the tools at our disposal, we can at least mitigate the damage, if not stop some schemes completely.

Editorial standards