HOW TO DETECT TERRORIST FINANCING USING ARTIFICIAL INTELLIGENCE

TERRORISTS ARE UNDERMINING THE INTEGRITY OF THE FINANCIAL SYSTEMS BY TRYING TO MOVE ILLICIT FUNDS WITHOUT BEING DETECTED. THE KEY CHALLENGE FOR FINANCIAL INSTITUTIONS IS TO IDENTIFY THESE FUNDS AND BLOCK THOSE PAYMENTS IN ORDER TO FULFILL THE REQUIREMENTS POSED BY COMPLIANCE REGULATIONS.

At the same time, banks and payment service providers have to optimize business processes, control their costs and manage human resources to handle the fast-growing volumes of digital payments on a day-to-day basis. The complexity of financial systems, new payment methods and the continuing change in sanction and terrorist finance regulations are increasingly pushing the limits. Recent attacks and the threats posed by the terrorist organizations globally leads to two questions: Would it possible to follow the trail of money from individuals related to terrorism? And if yes, how?

Counterterrorism sanctions are aimed at avoiding transactions directly linked to sanctioned countries, entities, and individuals. What about illicit funds from/to people who are not listed? How would you identify those?

Most automated transaction monitoring systems can identify transactions that are related to terrorism financing by using watch lists such as from the EU or OFAC. However, to find unknown connections, those who are not listed is a different story. To identify the “unknown” financier of terrorism you need to use a different search strategy for detection. For example, using other relevant information from other customer channels combined with your data could help an institution to better identify suspicious behavior.

Let’s have a closer look at one of the real case examples. It is about an individual related to a terrorist organization, but not registered on a list. How does a multi-channel monitoring approach help alert you and identify red flags?

A 20-year-old female student during a period of 6 months receives money transfers from a Non-Profit Organization (NPO) located in Jordan. After she collected enough money, she bought an airline ticket and left for Turkey. During the following month, several cash withdrawals from an ATM occurred close to the borders of the conflict zone with Syria. Then after seeing no account activity for six months, online banking activities occurred from an IP address close to a conflict area again, but this time in Jordan.

Is it something we can consider as suspicious? Yes, there are a number of red flags that we could classify as suspicious behavior:

• The transactions didn’t match the normal customer profile, such as occupation (student)
• It is atypical to see such transactions between a student and the NPO
• There were unusual cash withdrawals in the high-risk zones
• The age and gender of a person can be considered and flagged when it is combined with suspicious behavior. It can be an indicator of the potential support of a terrorist organization. According to recent study, especially young women between the ages of 16-24 have travelled to Syria to join ISIS
• No activity on the account for six months. Transaction dormancy could be the result of terrorist training or engagement in terrorist activities.

An interesting case below shows NPOs involvement in the terrorist activities:

“A client was receiving donations of money in his account in Switzerland from different people located in Germany. He informed the bank that he was using his private Swiss banking account for collecting donations because he could not open an account for his charity in Germany due to legal restrictions. The donations were meant to be withdrawn in cash and brought personally to Tanzania to build a fountain. According to the bank statements, different reasons were declared by the donators: “Donation Africa Fountain”, “Donation Streetwork”, “,Tansania Orphanage”, “Mosque Building”, “Koran School” etc. Media reported that the NPO “Africa Fountain” was closely located to an area controlled by extremists related to terrorism.”

The case above is a real example. The methods used for terrorist financing are in many cases identifiable, as well as the way the money is moving through the banking system.

There are several possibilities to enhance your risk assessment related to terrorism financing:

1. AI Technology: It’s an optimal method in preventing terrorist funding because it is an extension of the knowledge-based systems, such as machine learning. Machines will help to analyze many data connections and find the patterns not always noticeable to humans.
2. Data: The more quality data you have, the easier it is to identify anomalies. While you may know what is suspicious, the question is: how do you implement this in order to flag this type of behavior?”. Look for multi-channel systems, which can analyze not only big data but also transactions from different channels. It will help you to find abnormal behavior.
3. Expert knowledge: Make sure you that AI and data are integrated together with the help of experts in the AI field. Data scientists with several years’ experience working on AI algorithms will bring significant benefits to an organization. They will help configure the system so that it begins learning and identifying behavior without human oversight.

Financial institutions are playing an essential role in the chain of identification of terrorist financing. With the right tools, it is possible to reduce potential vulnerabilities. It is best to combine quality data, advanced AI technology, and expert knowledge to achieve the best results.

If you want to learn more about AI Technology, please contact our Compliance & Fraud experts at INFORM GmbH.