Fraud is a costly affair for any institution. A recent study revealed that every dollar of fraud lost last year to US financial firms represented $4.00 in costs. These losses span liability transaction costs, and expenses, interest, legal, labor, investigation, and external recovery fees. Furthermore, PwC’s Global Economic Survey 2022 claims that while the proportion of organizations experiencing fraud has stayed steady in recent years, there has been a marked increase in externally perpetrated fraud executed by hackers and organized crime units. The only effective way to combat hacking and other technology-centric fraud is for financial organizations to arm themselves with the newest technologies and implement robust fraud protection practices and policies. Here are the latest technologies in the 2022 risk-management tool kits of successful financial institutions.
Onboarding Client Risk Reduction
Fraudulent actions occur across all stages of the customer journey, from onboarding to final transactions. The potential to catch swindlers during the initial process of client identification and verification presents banks with a unique opportunity to pinpoint perpetrators and prevent further corruption of the system. KYC, or Know Your Customer, is a set of standard protocols developed to assist financial institutions in verifying a customer’s identity. Unfortunately, when it comes to the banking sector, the implementation of KYC has not always kept pace with advances in customer fraud, such as the use of Artificial Intelligence deep fake technologies. Only Artificial Intelligence, or AI, can competently assess deep fake imagery, as well as examine other unique biometric data.
The banking sector had previously been sluggish in its adoption of the latest AI technologies, which are made possible through access to both proprietary and open-source software solutions. This is changing, and in 2022 more banks are investing in transforming both their IT attitudes and infrastructures. KYC open source is fast becoming a more popular choice than static proprietary versions of customer authentication, as it’s more evolved. In harmony with this shift in perspective, banks are also leaning into other related methodologies such as Open APIs, DevOps, microservices, Agile, and the Cloud. This of course requires hiring employees with advanced education in Artificial Intelligence, development, and engineering. The long term benefits of these investments are clear for the banking industry.
Money Laundering Risk Remedy
Money laundering often used to disguise the proceeds of organized crime syndicates, is estimated to involve 2 to 5 percent of annual global GDP, with financial institutions the site of a significant proportion of this fraudulent activity. Banks are increasingly adopting new anti-money laundering technologies, or AMLs, that is based on Artificial Intelligence, and its divisions of Machine Learning, or ML, and Natural Language Processing. In the past there was a heavy reliance on frontline humans to identify and respond to suspicious financial transactions, resulting in often sluggish and ineffective AML strategies. In 2022 more banks are relinquishing their hesitancy towards AI AML measures, and coming on board with the latest technologies. An example of the recent successful adoption of AI AML can be found at SAS, a Tier 2 regional US bank. Recent SAS reporting shows that AI AML models have resulted in a reduced alert volume of 55%, and an increase in suspicious activity report (SAR) yields of 25%.
ML on Insider Alert
Not all fraud in the financial sector comes from outside, with insider actors also responsible for a large and growing chunk of the deception. Insiders may be direct bank employees, contractors, former employees, or other affiliated persons. These kinds of fraud can be trickier to unmask as the perpetrators benefit from their intimate knowledge of the institutional infrastructure, as well as their privileged access to personal customer data. In many cases, the internal fraudster will steal funds and transfer them from a customer’s account to a prepaid card, or other accounts that typically require little KYC compliance to set up.
In 2022, ML, a subset of AI, represents the latest in technology assisting in the identification of insider fraud. Investigative processes into potential insider fraud have often suffered due to their time consumption, high incidence of false alerts, lack of specific insights around customer accounts and entities, and the overwhelming volumes of data they throw up. ML focuses on data analysis methods and is more capable than AI alone of interrogating large amounts of data for specific behaviors and purposes.
Recent advancements in the interpretability and explainability of both Artificial Intelligence and the related sphere of Machine Learning have increased their value as predictive tools in the fight against financial fraud. Banks and other institutions in the sector are continuing to increase their adoption of AI and ML, enshrining the universal principle that prevention is better, and in this case often cheaper, than a cure.