Artificial Intelligence Identifies When Bank’s Internal Controls Are Failing for AML and Other Regulatory Requirements

The Financial Crimes Enforcement Network (FinCEN) recently assessed a $7 million civil monetary penalty (CMP) against a California-based bank for willful violations of several provisions of the Bank Secrecy Act (BSA).  Simultaneously, the bank’s primary regulator, the Office of the Comptroller of the Currency (OCC) assessed a $1 million CMP against the bank based on the OCC’s discovery that the bank was deficient in its compliance to a previously established consent order.

The bank was cited for failing to establish and implement an adequate anti-money laundering (AML) program, failing to conduct required due diligence of its foreign correspondent customers, and to detect and report suspicious activity.  FinCEN stated that the bank “allowed billions of dollars to flow through the U.S. financial system without effective monitoring to adequately detect and report suspicious activity.”

Many of the bank’s customers were located in high-risk jurisdictions and were related to money service businesses (MSBs).  FinCEN indicated that a large number of the MSB-related transactions were conducted through MSBs that were “owned or managed by bank insiders who encouraged staff to process these transactions without question or face potential dismissal or retaliation. Bank insiders directly interfered with the BSA staff’s attempts to investigate suspicious activity related to these insider owned accounts.”

With millions of dollars in penalties, this incident, and the Wells Fargo account fraud scandal in 2016, begs the question of how effective internal controls are in financial institutions.

A lack of appropriate controls puts stakeholders at risk and, as in this case, allows money laundering to go unchecked. This reinforces public and regulatory sentiment that banks are not doing enough to address this solvable problem and further bankrolls criminal activity such as illegal drug trade, human trafficking and terrorism.

There’s no reason for today’s financial institutions to have an inefficient AML program.  The proven capabilities of data science to expose suspicious entities and transactions not only improves regulatory compliance, but also informs the decision-making of senior management. Specific data science, AI and machine learning techniques that would have indicated anomalies worthy of further investigation include, but are not limited to:

• Collaborative filtering: capable of finding transactions with missing, matching and/or odd information
• Feature matching: utilized to identify transactions below a specific monetary threshold
• Fuzzy logic: used to find data matches with slight changes to names or addresses
• Cluster analysis: can detect abnormalities in transactions benefiting a single person or entity
• Time series analysis: detects transactions benefiting a person or entity over an extended period
• Focused keyword searches: ability to dynamically monitor, screen and filter transactions based on keywords from high-risk AML, counter-terrorism financing (CTF) and financial crimes typologies.
• Ability to learn from an AI-identified suspicious activity to enhance transaction monitoring and KYC platforms

The alarm has sounded in the compliance world and a modern AI-enhanced machine learning solution is the first responder.  AI-based solutions can prevent, detect and assist to report suspicious events, such as rogue employee activity or insiders with access to banking infrastructure, while enhancing customer due diligence/KYC.