News

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. Vendors have claimed that compliance programs upgraded with machine learning – a form of artificial intelligence, or AI, that relies on algorithms to spot patterns and refine future decision-making based on the results – will reduce "false positive" transactional alerts while identifying more nuanced suspicious activity that would typically escape detection. Minneapolis-based U.S. Bank unveiled a monitoring system last year with "machine-learning models" that can adjust their own thresholds for...