Despite major investments in compliance and artificial intelligence (AI), financial institutions continue to face major challenges in detecting and preventing illicit financial flows.
Chinese money laundering networks (CMLNs) alone have moved hundreds of billions of dollars through global systems, fuelling drug trafficking, human exploitation, and other criminal enterprises, claims Consilient.
According to the U.S. Department of the Treasury’s Financial Crimes Enforcement Network (FinCEN), a recent Financial Trend Analysis reviewed more than 137,000 Bank Secrecy Act filings between 2020 and 2024, uncovering over $300bn in suspicious transactions. The agency has urged institutions to stay alert to activity linked to these sophisticated laundering schemes, which intertwine with legitimate business operations such as real estate, luxury goods, and healthcare services.
The regulatory risks are steep. In one notable 2024 case, a major institution paid over $3bn and admitted criminal liability after failing to identify laundering activity, including a customer funnelling $470m in drug proceeds through its accounts. FinCEN and other regulators are now encouraging banks to adopt innovative anti-money laundering (AML) tools, such as federated AI, to strengthen oversight and reporting standards.
Chinese money laundering networks have evolved as an intricate system connecting Chinese nationals seeking U.S. dollars with Mexican cartels in need of peso liquidity. Restrictions in both countries—China’s cap on foreign currency conversion and Mexico’s limits on dollar deposits—have created an underground market that thrives on mutual demand. CMLNs operate through informal transfer systems, trade-based schemes, and coordinated “mirror” transactions between nations, often using complicit merchants and money mules to disguise the flow of cash into legitimate assets.
For example, cartel affiliates in the U.S. provide cash to CMLN agents, who transfer equivalent pesos to Mexico-based cartel accounts. The dollars are then sold to Chinese buyers who move yuan into domestic bank accounts, while their agents in the U.S. deposit the cash into local institutions. This process not only launders proceeds from narcotics but also facilitates fraud, human trafficking, and other illicit trades.
To counter these threats, FinCEN has outlined 18 key red flags for financial institutions. Indicators include large cash deposits inconsistent with a customer’s occupation, same-day wire transfers abroad, suspect use of Chinese passports, high-value cashier’s checks for property purchases, and rapid inflows or outflows from new accounts. Firms must file Suspicious Activity Reports (SARs) whenever such transactions arise and establish rigorous, risk-based due diligence frameworks to identify and mitigate exposure.
Federated AI offers a transformative approach to AML collaboration. Rather than pooling sensitive client data, participating institutions each train their models on proprietary datasets within secure environments. The aggregated learnings—stripped of customer identifiers—are then combined into a central model, enhancing detection accuracy across institutions.
For AML use cases, this technology allows financial institutions to recognise broader patterns of behaviour that single-entity systems might overlook, such as interconnected mule accounts or coordinated cross-bank activity. The resulting “collective defence” network enables enhanced detection capabilities while ensuring privacy compliance and data sovereignty.
As money launderers exploit fragmented regulatory systems, federated learning provides a scalable, privacy-preserving framework for global collaboration. Financial institutions that embrace such technologies not only bolster their defences against criminal activity but also demonstrate to regulators their proactive commitment to compliance innovation and due diligence.
Find more on RegTech Analyst.
Copyright © 2025 FinTech Global
