Creating, implementing, and maintaining an effective Anti-Money Laundering (AML) program has always been challenging, but in today’s increasingly digital, dynamic environment the task is more arduous. Given the complexity of the current AML landscape, quality data is often viewed as more valuable than gold or oil by financial institutions and non-bank financial institutions. Artificial intelligence (AI) and machine learning solutions drive compliance and audit teams to peak performance and efficiently, and effectively analyze millions of transactions to find hidden risk and solve root cause analysis issues.
The increasing AML program scrutiny is exemplified by the Securities and Exchange Commission’s (SEC) Office of Inspections and Examinations announcement of its 2018 examination priorities, which touched on AML. The priorities are published annually in an effort to improve compliance, prevent fraud, monitor risk, and inform policy.
The SEC’s 2018 priorities include: Compliance and risks in critical market infrastructure; matters of importance to retail investors, including seniors and those saving for retirement; FINRA and MSRB; cybersecurity; and AML. The SEC indicated stated that “examiners will review for compliance with applicable anti-money laundering requirements, including whether firms are appropriately adapting their AML programs to address their regulatory obligations.”
The required pillars of an effective AML program are well known:
- Having written policies and procedures;
- Designating an AML compliance officer;
- Conducting independent testing of the institution’s AML program;
- Implementation of an adequate employee training program, and;
- Establishment of a risk-based, customer due-diligence (CDD) procedure (effective May 11, 2018)
A recent example of a failed compliance program was announced by the Department of Justice (DOJ) in which it charged the fifth largest bank in the United States for its failure to maintain an adequate AML program and failing to file timely suspicious activity reports. According to the U.S. attorney who headed up the investigation, the bank operated its AML program “on the cheap” by reducing the headcount of compliance analysts and other resources as well as capping the number of transactions subject to AML review. Additionally, more than 5,000 currency transaction reports were incomplete or contained inaccurate information, which impeded law enforcement’s ability to track potential financial criminals.
As a result, the bank failed to detect and investigate large numbers of suspicious transactions and has thus agreed to pay a penalty of $613 million, one of the largest settlements to date. This case underscores the importance of maintaining an effective AML program which should be driven by financial crime risk – not by the bank’s compliance resources or staffing concerns.
Established AI solutions can assist compliance and audit leaders with independent testing of AML programs and to perfect a risk-based CDD program. By supplementing the traditional review of transactional and customer data, financial institutions and non-bank financial institutions can greatly enhance their reputational, operational, and legal risk exposure by deploying advanced data science and AI techniques.
Artificial intelligence, supplemented by entity resolution and verification, UBO analysis, deep Web analytics, NLP (Natural Language Processing), Web scraping, network analysis, and volumes and values analysis, will assist compliance and audits teams to drive efficiency while solving their daily dynamic risk challenges.