The US Anti-Money Laundering Act 2020 supports the deployment of technological innovation and machine learning

US Anti-Money Laundering Act
The Anti-Money Laundering Act 2020 (AMLA) is a major watershed in the AML legislation of the US. It was passed by the Congress on January 1, 2021.

The objective of the AMLA is to increase the effectiveness of the AML system by improving communication, data sharing and data processing between the different entities of the AML network. We focus here on the section of the AMLA entitled « Modernizing the AML/CFT System, » in which we discern several key core principles: (1) law enforcement feedback, (2) use of new technologies, including AI, (3) review of reporting requirements, and (4) systematic evaluation of the effectiveness of deployed provisions.

An emphasis on law enforcement feedback

The AMLA calls for increased feedback from FinCEN (the US Financial Investigation Unit) and the Department of Justice to reporting institutions and financial regulators. The emphasis on feedback is articulated in many provisions in the law: FinCEN must organize periodic ‘feedback sessions’ with reporting entities, must disclose periodically information on the SARs filed that proved useful to Federal or State criminal or civil law enforcement agencies, or shall provide banks and regulators with threat patterns (typologies of money-laundering). For its part, the Attorney General must submit an annual report “that contains statistics, metrics, and other information on the use of reported data […] that lead to further procedures by law enforcement agencies”.

Technological innovation support

Another important highlight of the law is on the use of “innovative approaches such as machine learning or other enhanced data analytics processes” to reinforce financial institutions’ and FinCEN’s crime detection capabilities. As previously mentioned, the law acknowledges the importance of feedback to improve data sharing and reliability, specifically  for potential use in algorithms. It also provides for the creation of a ‘Subcommittee on Innovation and Technology’ to advise on means to support technological innovation, composed notably of “Innovation Officers” from FinCEN and each Federal functional regulator.

Review of reporting requirements and assessment of the means employed

The law provides for the review of the reporting requirements based on cost/benefit analyses of “the means or form in which suspicious transaction reports are transmitted”. Building on this topic, the AMLA also requires FinCEN to implement, as appropriate, “streamlined data and real-time reporting”.

One notable change in the reporting requirements is the review of the thresholds for SARs and CTRs, which haven’t been revised since 1996 and 1970 respectively, and currently stand at $5000 and $10 000.

Impact assessment of the new provisions is also expected for innovative approaches using machine learning. These tools must be analysed using a risk-based approach in order to « prudently assess and monitor the effectiveness of their implementation. » More generally, the Treasury Secretary is required to « analyze the impact of financial technologies on financial crimes. »

Other provisions for sharing and processing of critical information include the enhanced protection of whistleblowers and the implementation by FinCEN of a beneficial ownership registration database.


By Astrid Bertrand, Télécom Paris – Institut Polytechnique de Paris