The US Department of the Treasury’s Financial Crimes Enforcement Network (FinCEN) published a Notice of Proposed Rulemaking (NPRM) on 7 April, setting out sweeping reforms to how financial institutions (FIs) must structure their anti-money laundering and countering the financing of terrorism (AML/CFT) programmes.
According to Workfusion, the proposals mark the most significant overhaul of AML/CFT requirements in roughly 25 years — and they send an unmistakable signal to the industry: FIs that deploy artificial intelligence in their compliance operations are likely to be viewed favourably by regulators.
Workfusion recently detailed how FinCEN’s new AML/CFT program rules dnderscore financial institutions’ need to use AI.
The NPRM fact sheet makes this explicit, noting that FinCEN’s director would consider “whether the bank is employing innovative tools such as artificial intelligence that demonstrate the effectiveness of the bank’s AML/CFT program” when deciding whether to pursue enforcement or supervisory action.
Three reasons regulators are embracing AI for AML/CFT
Regulatory enthusiasm for AI in compliance is not emerging in a vacuum — there are structural forces driving it. First, criminal networks are now exploiting AI to execute fraud, money laundering, and other financial crimes at a scale and velocity that manual processes simply cannot match. For banks, FinTechs, and other FIs, deploying AI and automation is the only credible path to keeping pace with increasingly sophisticated bad actors.
Second, governments are engaged in an active competition for global financial market influence. This has pushed regulatory bodies to modernise rapidly and develop a working understanding of emerging technologies. Regulators that succeed in doing so will be better placed to encourage innovation at financial services firms whilst simultaneously bolstering defences against financial crime.
Third, AI agents bring meaningful crime prevention capabilities to AML/CFT compliance with relatively contained risk. The processes involved are well-defined and can be isolated, making them well suited to automation. WorkFusion’s FI customers already use AI agents to analyse and action millions of alerts daily across areas including payment screening, name screening, adverse media monitoring, enhanced due diligence (EDD), AML transaction monitoring, and fraud. When FIs underpin their compliance operations with AI in this way, regulators can feel greater confidence that programmes are calibrated to meet modern threats at the required scale and speed.
A sharper focus on high-risk customers
Another central theme of the NPRM is the Treasury’s intention to reduce the burden of unnecessary compliance work. The proposals explicitly call for “empowering financial institutions to direct more attention and resources toward higher-risk customers and activities rather than toward lower-risk customers and activities.” In practice, this means FIs must find smarter ways to surface the riskiest customers and transactions from the constant noise generated by high-volume screening tools.
This is where investigative AI agents become particularly relevant. WorkFusion’s AI agent Edward is designed to perform the deeper analytical work required for EDD on high-risk customers. Unlike simpler automated tools, Edward applies contextual reasoning to complex data sets — enabling it to assess a customer’s risk profile in the round rather than simply flagging surface-level anomalies.
Consider a US-based manufacturer classified as high risk by its bank due to its international supplier relationships, use of multiple shipping services, and frequent cross-border payments. Edward would conduct an annual account review, examining factors such as the types of products the company manufactures, the plausibility of its supplier relationships, ownership information across all counterparties, and the headquarters and branch locations of business partners, among other considerations. By synthesising this breadth of information and reasoning through it contextually, Edward is reported to save human investigators between one and three hours per case on average — translating to faster response times and reduced loss exposure for FIs serving high-risk clients.
With FinCEN’s reforms pointing unmistakably towards AI as a marker of compliance programme maturity, the question for financial institutions is less whether to adopt AI-driven AML/CFT tools, and more how quickly they can do so.
Read the full Workfusion post here.
Copyright © 2026 FinTech Global
