Improve fraud detection using AI and ML


The threat financial crime poses to businesses has never been greater.

Almost three-quarters of compliance professionals in a survey said they filed more than Suspicious activity reports related to financial crimes in 2020 than in 2019, and 69% said better fraud detection would have the most impact on reducing losses. More than half also cited cybersecurity as their # 1 vulnerability.

Financial cybercrime in particular has increased alongside the accelerated shift to remote working. The FBI received nearly 800,000 cybercriminality reports in 2020 – a 69% increase from 2019. Financial crimes accounted for the majority of cases and were also the costliest, with victims losing $ 1.8 billion just to compromised emails professionals (BEC). Other research indicated that e-commerce merchants would lose more than $ 20 billion to the benefit of payment fraud this year – 18% more than in 2020 – and this trend to get worse in the years to come.

New PYMNTS search have shown that banks and other financial institutions (FIs) are more frequently implementing artificial intelligence (AI) and machine learning (ML) in the fight against fraud. The following Deep Dive examines the importance of AI and ML technologies in fraud monitoring and prevention, including the main advantages they offer over traditional methods. It also highlights the gap in the adoption of these tools between large and small businesses and the perceived barriers that may prevent more businesses from getting started.

AI and ML anti-fraud capabilities

Online fraud is one of the biggest challenges businesses face in the digital space. A 2020 PwC survey of more than 5,000 companies found that 47% had encountered fraud attempts over the past 24 months, with known fraud losses totaling $ 42 billion. Association of Certified Fraud Examiners Global Study of Over 2,500 Fraud Cases Finds Businesses Suffered median damage of $ 125,000 per case in 2020, 25% losing more than $ 600,000.

FIs have traditionally used rule-based systems as well as a manual review for fraud detection. The increasing sophistication of fraudsters and the difficulty of creating rules for each abnormal transaction challenge this method. False positives – reporting legitimate customers as fraudulent – and fraud undetected due to high data volumes are two of the most difficult issues plaguing rules-based systems. AI and ML, on the other hand, rely on the much more efficient principle of detect deviations of standard activity.

ML specializes in detecting outliers in large data sets and in “training” and adapting to new input models. Banks can deploy technology for supervised and unsupervised learning, which can strengthen their fraud prevention efforts in a number of ways. Supervised learning enables FIs to receive real-time information about their fraud analyzes, helping them to better tailor their solutions to eliminate false positives based on the experiences of fraud investigators. Unsupervised learning, on the other hand, allows banks to leverage ML to eliminate potential fraud scenarios that are not described in the existing analytical framework.

AI and ML systems thus excel in fraud prevention because they can identify subtle trends in the ever-changing approaches of sophisticated cybercriminals as they increasingly use AI themselves. This explains why more and more industry players, governments and auditors are embracing AI and ML for fraud prevention in place of their old rule-based systems.

Benefits and Barriers to AI and ML Adoption

The main players in payments such as American Express and Visa have deployed highly advanced AI-based systems for fraud and credit risk monitoring and continue to invest in their AI programs. Synchrony has also achieved an accuracy rate of over 90% with its AI anti-fraud ecosystem. A PYMNTS study surveyed financial institutions in the United States about their use of AI and other advanced technologies, and the results confirm rapid growth in adoption of AI systems, which roughly tripled between 2018 and 2021, from 5.5% to 16% of respondents. This aligns with broader industry trends and, while still weak, suggests that most businesses could be using AI solutions within a few years.

FIs surveyed who had previously adopted AI identified the benefits directly or indirectly related to fraud surveillance and prevention as the top five benefits of the technology.

Eighty-one percent said they were alerted to fraud before it happened, 75% cited reduction in false positives and 56% reported reduction in payment fraud as a key result of their systems of AI, all of which are essential anti-fraud functions. The other two main benefits – improved operational efficiency (81%) and improved customer satisfaction and experience (63%) – are also indirectly related to effective fraud prevention, as a Accurate and transparent AI-based system will increase efficiency and keep customer experience frictionless.

However, only the large FIs surveyed tended to have ever adopted AI systems. In total, 79% of FIs with more than $ 100 billion in assets said they used AI compared to only 4.5% in the $ 25-100 billion range, and none used it among those with less than $ 25 billion in assets.

Another PYMNTS study highlighted why some companies haven’t implemented AI systems. Seventy-two percent cited regulatory issues, making it the most common concern, followed by the complexity of AI (59%) and its higher data management costs (59%). These findings identify the top concerns – and perhaps the misperceptions – that AI and ML service providers should seek to address in their messages to potential customers.

However, many small businesses are considering starting to use AI. Ninety-nine percent of companies said they have already invested in AI systems (21%) or plan to be within three years (72%), most of these within 12 months (57 %).

These good results suggest that AI systems have reached a tipping point in terms of interest among FIs, although some are not implementing the technology as quickly as they would like. AI and ML offer major advantages over rules-based systems for banks and other institutions in the fight against fraud, including the rapid and accurate identification of fraud before it occurs, reduction of false positives, elimination of manual labor costs and improved customer experience. Businesses that want to stay competitive in a rapidly changing fraud landscape will want to invest in these systems right away.


Comments are closed.