As anti-money laundering (AML) compliance becomes ever more important, banks need to be able to demonstrate the effectiveness of their transaction monitoring systems. Unfortunately, manually evaluating these models can be slow and, if done incorrectly, expensive. To avoid this, more and more banks are turning to automated machine learning to improve tracking of suspicious transactions.
Under the OCC supervisory guidance, banks must comply with regulations and goals, which includes the criticality of analyzing the limitations and assumptions of models to produce appropriate changes. Accordingly, regulators expect banks to efficiently manage model risks via frequent evaluation, with many expecting bank models to be evaluated annually.
Further emphasizing the importance of AML, in 2022 the US Department of Justice (DoJ) announced that Chief Executive and Chief Compliance Officer certifications may soon be required for an effective compliance program. This demonstrates that the US is determined to reduce access to sanctioned countries and individuals, as well as cracking down on financial crime in general.
For smaller banks, manually evaluating models can be a burden. Thankfully, automated machine learning (AutoML) can help optimize this process and increase the efficiency of model evaluation. By adopting AutoML, banks can train models that learn good customer behaviors and recognize patterns of financial crime. This approach can improve banks’ ability to adapt to ever-changing AML regulations, giving them a greater chance of staying ahead of threats.
The Danske Bank is one example of a company that failed to maintain a sufficiently effective anti-money laundering program. For this failing, Danske Bank was recently fined over a billion dollars.
The Deputy Attorney General for the US Department of Justice (DOJ) Lisa Monaco remarked when announcing the fine that companies should invest in robust compliance programs. With added pressure from these ever-larger fines, automated machine learning provides an important security measure so that banks and FIs can keep up and continue to protect themselves and their customers.
