Thank you to technological advances, it is becoming increasingly convenient for consumers to carry out online transactions, such as online purchases and money transfers by clicking on a link or using a mobile phone number (authorised push payments). On the
other hand, the increasing acceptance of online transactions makes it easier for fraudsters to trick their victims and cause financial damage to financial institutions’ customers (which can amount to millions of dollars) and damage the reputation of financial
institutions due to the lack of effective anti-fraud measures.
To strengthen cybersecurity and minimise losses from fraud, regulators in various jurisdictions have started publishing guidelines for financial intermediaries to give them guidance on how to protect their customers from fraudulent activities.
The Monetary Board of Philippines issued Circular No. 1140 dated March 17, 2022 (https://www.bsp.gov.ph/Regulations/Issuances/2022/1140.pdf), which emphasises the importance
of a real-time fraud monitoring system (FMS) to protect customers from cyber fraud in electronic transactions. In addition, the regulator suggests linking or integrating both the FMS and AML systems to create a coherent and comprehensive system for the prevention
of financial crime.
The Hong Kong Monetary Authority (HKMA) has issued various circulars in 2023 to provide guidance to local financial institutions to further strengthen electronic banking (e-banking). The latest circular dated October 31, 2023 (https://www.hkma.gov.hk/media/eng/doc/key-information/guidelines-and-circular/2023/20231031e1.pdf)
also emphasises the importance of real-time fraud prevention. In addition, the HKMA emphasises the use of “dynamic” fraud monitoring rules that combine historical customer data and transaction patterns to counter the increasing sophistication of fraud tactics.
To summarise, an effective and efficient fraud monitoring solution should have the following features:
1. Real-time and the ability to handle the ever-increasing volume of online transactions within a reasonable timeframe (typically, the timeframe for the fraud monitoring process is around 100 ms)
2. The ability to build a customer's behaviour (profile) to screen out suspicious activity not only in monetary but also in non-monetary transactions.
3. Combination of data from online and offline channels (e.g. ATM withdrawals, in-branch activity, telephone banking, etc.) to identify suspicious activity.
4. Convergence of fraud and AML systems for a comprehensive financial crime prevention system.