Banks that use artificial intelligence to streamline their monitoring of transactions for anti-money laundering purposes must account for the possibility that the technology could yield discriminatory results and exacerbate de-risking, analysts told ACAMS moneylaundering.com. Regulatory concerns over AI's potential for fueling financial exclusion have largely focused on whether the technology could influence banks to unfairly withhold credit from certain customers, especially if, as the Office of the Comptroller of the Currency warned in October, such systems are "improperly trained or used with datasets that reflect biases or past discrimination practices." Despite that framing, the problem extends beyond credit: AML personnel...
Years after U.S. regulators first prodded banks to sharpen their anti-money laundering tools with innovative technology, sources told ACAMS moneylaundering.com that compliance staff can expect more pressure to adopt machine learning and other artificial intelligence-driven systems.
Global lenders running machine-learning programs alongside their current, more traditional transaction-filtering systems have garnered cautious praise from U.S. regulators, but numerous obstacles toward full implementation remain, sources told ACAMS moneylaundering.com.