The focus of a money-laundering probe into Denmark-based Danske Bank grew from $150 billion in suspicious transactions, as reported earlier this month, to $230 billion, according to the Wall Street Journal.
An internal anti-money laundering (AML) investigation has revealed that the bank’s small branch in Estonia was responsible for funneling the massive sum of money, which transferred funds from ex-Soviet states into countries such as China, Turkey, and Switzerland. The Wall Street Journal reports the sum is 26 times greater than total government spending in Estonia for 2015, the year many of the accounts were shuttered.
The threat of money-laundering fines is looming large in 2018. In the first half of 2018 alone, global fines totaled $1.7 billion for AML compliance failures — which is 85% of the $2 billion total amount of fines in the entire year 2017, according to Debevoise & Plimpton.
US prosecutors and regulators have been particularly aggressive in leveling fines, originating over $1 billion of the $1.7 billion in global fines for the first half of 2018. In addition to monetary damages, money-laundering fines can damage banks’ reputations, prompt criminal charges against bank employees, or in the most extreme cases, lead to access to US dollars being cut off altogether.
Illegal money flows from Russia are behind much of the rising money laundering threat. In addition to Danske Bank, two recent examples highlight this trend:
- Latvia ABLV bank was liquidated in February after the US Treasury cut off access to dollars due to money laundering, according to Forbes.
- Deutsche Bank was fined $630 million in 2017 for a stock-trading scheme that allowed $10 billion to illegally exit Russia, according to CNBC.
Artificial intelligence (AI) could help banks prevent money laundering efficiently. To stem rising fines, banks may need to grow their already high AML budgets, which exceeded $8 billion worldwide in 2017, according to WealthInsight.
One approach banks can use to limit rising costs is to employ machine learning— rather than bank employees — to sort suspicious transaction alerts into low-, medium-, and high-risk buckets, according to Accenture.
However, because human approval and review is deemed necessary by law for many AML processes, algorithmic solutions are likely to augment humans rather than outright replace their role in preventing illicit activity for the foreseeable future.
This is a subscriber-only story. To read the full article, simply click here to claim your deal and get access to all exclusive Business Insider PRIME content.
BI Intelligence Content Marketing