Businesses require a broad fraud prevention strategy. This means being proactive about reducing threats and protecting their income.
Traditionally, North America and Europe have been prime targets for eCommerce fraud, but it’s rapidly expanding in other regions, too. The recently released Identity Fraud Report 2024 from Sumsub, for example, reveals a rapid increase in fraud in the Asia-Pacific (APAC) region, calling for immediate solutions.
The report signals a rampant rise in identity fraud across the region. Overall, APAC saw a whopping 121% increase in identity fraud attacks between 2023 and 2024. Some countries were hit significantly more than others. Singapore, for example, saw a massive 207% surge in identity fraud compared to last year, while Thailand and Indonesia recorded a 206% and 201% year-on-year increase, respectively.
The report sheds light on five prominent forms of identity fraud in the APAC region. Many developing countries are regretfully ill-prepared to combat the rise in fraud. Indonesia’s fraud rate is the highest in the APAC region at 6.02%. Pakistan and Bangladesh follow closely with fraud rates at 4.28% and 4%, respectively. Conversely, only 1.66% of transactions in the United States are considered fraudulent.
What Does This Mean for Merchants Operating in the APAC Region?
In short, it means merchants operating in Asia will probably encounter a rise in identity fraud attacks. Specifically, those eCommerce merchants located in the region could be highly vulnerable.
That said, chargeback fraud is a particularly significant issue, too. Sumsub provides additional data supporting the claim that chargeback fraud is increasing. According to the study, 56% of merchants worldwide underwent chargeback fraud in 2024; a problem that is notably pronounced in the APAC region.
More worryingly, merchants struggling with fraudulent chargebacks are likely to lose the battle. As Visa points out, these merchants have a dispute win rate of less than 20%. Sellers will ultimately recover just $156 out of every $1,000 disputed. This figure emphasizes the challenges that merchants encounter in responding to fraudulent dispute claims.
Interestingly, fraudsters are increasingly using AI as an affordable and scalable way to commit fraud. Deepfake attacks, which grew four-fold globally between 2023 and 2024, are exclusively created using artificial intelligence (AI) technology. But, while this is a significant concern, it also poses a critical question: could we leverage the same technology to counteract fraud and chargebacks?
Using AI Fraud Detection to Fight AI-Enabled Fraud
Machine learning models, which are essentially a subset of AI, are incredibly effective at detecting fraud. They do this by using historical data to train algorithms on the patterns of fraudulent and legitimate transactions. The model can then apply those learnings to any new incoming transaction, identifying whether it is likely to be fraudulent or not.
This method, known as supervised learning, has been proven highly effective in decreasing false positives and fraud losses for merchants. This kind of technology could significantly benefit merchants operating not just in the APAC region, but the world at large.
As for deepfakes, machine learning offers possible solutions as well. AI-enabled tools can analyze visual, audio, video lighting, and even the shape of hands for inconsistencies. Similarly, advanced machine learning algorithms can work in conjunction with geolocation and device fingerprinting technologies to halt fraud networks. Furthermore, AI technologies can help identify suspicious patterns or anomalies in customer behavior that may indicate fraudulent activity.
And, as technology continues to advance, so do the capabilities of AI. In the future, we could see more sophisticated use of machine learning models to combat fraud. This could include combining supervised learning with unsupervised learning techniques to better detect emerging fraud trends and attack methods.
Identity fraud is a complex and varied threat. A single solution can’t curtail all fraudulent attacks. Instead, merchants need a multi-dimensional strategy that addresses diverse pre- and post-transaction threats from multiple perspectives.
Implementing a comprehensive fraud prevention strategy is vital in the current eCommerce environment. The trend towards wider adoption of AI and machine learning technology isn’t likely to reverse. So adapting is critical. Businesses that prioritize proactive fraud mitigation will undoubtedly have a better shot at maintaining long-term profitability.
