How we helped the bank Protect Themselves Against Fraudsters: Implementing an Anti-Fraud System

How we helped the bank Protect Themselves Against Fraudsters: Implementing an Anti-Fraud System

How often do we hear stories about friends or acquaintances having funds fraudulently withdrawn from their accounts? Chances are, everyone knows someone who has, in one way or another, fallen victim to fraud.

Such incidents greatly affected the bank’s reputation, even though, in reality, the user’s carelessness was to blame.

According to research by Javelin Strategy & Research, in 2022, $32.2 billion in fraud was committed against bank accounts in the United States. This accounts for about 0.7% of the total transaction volume.

The most common types of banking account fraud:

• Phishing – involves sending emails or text messages that appear to be official communications from the bank. These messages often contain links or attachments that, if clicked or opened, can infect the user’s device with malware or redirect them to a fraudulent website resembling the bank’s official site.
• Social Engineering – the form of fraud employs psychological manipulation to deceive individuals into revealing their personal information or performing actions that compromise their bank accounts. It often involves direct interaction with the victim, either through phone calls, emails, or even in-person contact.
• Hacking  – entails unauthorized penetration of the bank’s computer systems to access and steal customer data. Hackers exploit vulnerabilities in the bank’s cybersecurity defenses to gain access to sensitive information.
• Gen AI Frauds – a newly emerging threat, this type of fraud involves using sophisticated generative artificial intelligence (AI) technologies to create highly convincing fake content. This could include AI-generated emails, voice messages, or even deepfake videos that mimic legitimate bank communications or personnel. Gen AI frauds are particularly challenging to detect because of their high realism and accuracy in mimicking genuine interactions.

Social fraud is one of the key issues banks struggle to cope with, and it is not easy to combat.

When we began analyzing the operations of our client’s bank, we discovered a concerning trend: the institution was encountering around 200 cases of social engineering fraud each month. In these cases, imposters posing as bank representatives were adept at phishing for customers’ personal information, which they then exploited for fraudulent activities involving the customers’ accounts.

Implementation of the Antifraud System

The initial phase of implementing the antifraud system in our client’s bank involved developing a core engine, followed by iterative enhancements. Initially, a blacklist approach was adopted, where all devices previously associated with fraudulent activities were barred from conducting transactions.

Subsequently, the system evolved to incorporate more nuanced criteria for blacklisting, akin to a traffic light system. Under this system, a “red” status indicated that while transactions were possible, they required additional verification through the contact center.

An application was developed for the contact center, where all pending transactions were flagged for review. Contact center employees would then reach out to clients, verifying whether they intended to execute the flagged transactions, essentially introducing a layer of manual verification for certain transactions.

As the solution matured, more sophisticated security measures like biometric verification, including face ID, were integrated. All data collected through these security measures were stored on the bank’s servers, ensuring that sensitive information remained within the bank’s secure environment.

Wins and Results

As the system evolved, we were able to drastically reduce fraud instances by more than 95%, bringing the number down from 200 cases to just 2-3. This remarkable reduction in fraud cases was a testament to the effectiveness of our antifraud system.

In evaluating the efficiency of this system, two key metrics were crucial: the total number of fraud cases and the sum of financial losses incurred. These metrics provided a clear indication of the system’s impact in safeguarding the bank against fraud and significantly minimizing financial losses. The substantial decrease in fraud instances not only enhanced the security of the bank’s operations but also reinforced customer trust and confidence in the bank’s ability to protect its financial assets.