Today’s technology has changed the way people manage their finances. Nowadays, many financial activities can be done using smartphones, such as paying electricity or water bills, shopping online through Shopee or Lazada and transferring money using DuitNow or online banking. Mobile financial services have made banking more accessible, especially for communities with limited access to traditional banking facilities. This development has brought better convenience, faster transactions and wider financial inclusion in many regions. Although digital finance brings many advantages, it also introduces new challenges. As financial activities increasingly move online, security aspects become more complex. Therefore, the financial system needs to continue to evolve to protect consumers and maintain trust in digital transactions.

One of the growing concerns in digital finance is financial crime, especially money laundering. According to Soltani et al. (2016), money laundering is the process of making illegally obtained money appear legitimate by hiding its true source. It usually involves moving funds through a series of transactions so that the original source becomes difficult to trace. In today’s digital financial environment, this process has become even harder to detect. Transactions can happen within seconds, across multiple platforms and sometimes across different countries. Because of this speed and complexity, identifying suspicious activities is no longer straightforward.

For years, financial institutions have relied on rule-based monitoring systems. These systems use pre-established rules, such as identifying unusually large transactions or activity in a particular location. While these methods have helped maintain financial security, they can struggle to keep up with evolving financial behavior. Artificial Intelligence (AI) is emerging as a tool for improving digital financial security. Unlike traditional rules-based systems, AI models learn patterns from training data, allowing them to adapt to changing transaction patterns and identify unusual behavior more accurately. This approach can improve monitoring efficiency, reduce reliance on manual processes and help detect potential risks more quickly. For example, deep learning techniques are well-suited to analyzing complex transaction sequences, as they are able to examine transaction patterns over time, detect anomalies and highlight activities that may require further investigation.

As illustrated in Figure 1, AI-based financial monitoring follows a structured process designed to handle large volumes of transaction data more efficiently. This process usually begins with the collection of financial transaction data from digital platforms such as online banking systems (Maybank2u and CIMB Clicks), payment gateways (iPay88 and FPX) or mobile financial applications (Touch ‘n Go eWallet and GrabPay). This data is then prepared through steps such as timestamp processing, account ID encoding, categorical feature encoding and selecting relevant information so that meaningful analysis can be carried out.

Once the data has been prepared, AI models begin examining transaction patterns from different angles. Rather than looking at transactions one by one, the system observes how financial activities change over time and how different accounts are connected to each other. This wider view allows the system to see patterns that might otherwise go unnoticed. For example, a sudden series of rapid transfers between multiple accounts, repeated small transactions within a short period, or transactions occurring at unusual hours may signal potential risk. Individually, these actions might appear normal. However, when viewed together as part of a broader pattern, they may indicate behaviour that deserves closer attention.

This structured analytical process enables financial institutions to monitor transactions more systematically and efficiently. Instead of relying heavily on manual reviews, AI enables faster detection of unusual activity and helps prioritise cases that require closer attention. As financial ecosystems continue to digitalise and integrate across platforms, intelligent monitoring has shifted from being optional to mandatory. It supports stronger security controls, enhances transparency and reinforces public trust in the financial system.

Financial crime not only affects banks but also everyday users. Victims may lose money, have their identity misused and face financial problems in the long term. Therefore, improving how these activities are detected is not just about technology but also about protecting people.

In the future, the integration of AI with financial systems is expected to become more advanced, allowing real-time monitoring and stronger protection mechanisms. Continuous improvement in AI models, along with collaboration between financial institutions and regulatory bodies, will contribute to a more secure and reliable digital financial environment.

By Dr. Wan Noor Hamiza Wan Ali, Senior Lecturer, Faculty of Artificial Intelligence (FAI) Universiti Teknologi Malaysia (UTM)