Artificial intelligence in anti-money laundering (AML) has reached a decisive moment, according to Napier AI. Regulators, banks and vendors are no longer debating whether AI belongs in financial crime compliance.
The real question is which jurisdiction will define the rules first. In a new analysis, Napier AI’s Dr Janet Bastiman argues that whoever establishes the first credible framework for explainable AI in AML will effectively hand the rest of the world its playbook.
Napier AI notes that AI in financial crime has shifted from tentative pilots to boardroom mandates. However, Bastiman believes the markets that lead will be those with the regulatory maturity to hold AI to account.
Rather than one global timeline, Napier AI identifies three regulatory postures. The “AI accelerators”, the UK and France, already have the sandboxes, risk-based supervision and public-private collaboration needed for AI-driven transaction monitoring and entity-centric risk scoring to mature safely. The “AI appreciators”, the USA, the EU and Singapore, are enthusiastic but are working through fragmented rulebooks or AI legislation not written with AML in mind. Finally, the “AI appraisers”, Australia, Canada and the UAE, favour auditability and explainability over speed. Paradoxically, this caution may result in them setting the toughest global standard.
Crucially, Napier AI points out that none of these jurisdictions mandates AI outright. Instead, regulators are converging on outcomes such as explainability, auditability and human oversight, leaving technology choices to institutions. The EU AI Act comes closest to a binding framework, classing transaction monitoring, KYC and sanctions screening as “high-risk” when AI is used. FinCEN in the USA is signalling that AI adoption can positively influence supervisory outcomes. Meanwhile, AUSTRAC in Australia is making AI a practical necessity by demanding detection standards that legacy rules-only systems can no longer meet.
Napier AI argues that regulators are approving specific, well-evidenced use cases rather than AI as a category. First-wave applications include transaction monitoring optimisation, alert prioritisation, customer risk scoring and Suspicious Activity Report drafting support. Network analytics, cross-institution intelligence sharing and continuous KYC are expected to follow. Fully autonomous decisioning for supposedly “low risk” transactions remains under heavy scrutiny, because defining low risk is itself a bias-sensitive judgement.
Napier AI is also sceptical about agentic AI as a standalone answer. An agent that triages alerts faster without improving detection quality simply moves risk around more efficiently while adding another layer that must be explainable and auditable. The firm believes regulators will instead trust AI built on embedded, jurisdiction-specific risk intelligence, shared typology knowledge and continuous validation.
For financial institutions, Napier AI’s message is clear. First movers should pilot explainable use cases now, fast followers can import proven approaches, and those in cautious markets should use this window to strengthen governance before the bar rises. According to Napier AI, the winners will be those with compliance-first foundations, not the most complex models.
For more, read the full story here.
Copyright © 2026 FinTech Global
