The future of financial crime detection will hinge less on how well individual firms manage risk in isolation, and more on how effectively the industry collaborates to spot it, according to Napier AI’s Dr Janet Bastiman.
For decades, anti-money laundering (AML) strategies have been built institution by institution. Each firm develops its own models, tunes its own rules and investigates its own alerts, largely alone. But financial crime ignores these boundaries entirely, moving across firms, jurisdictions and payment rails with growing speed. The result, Napier AI argues, is a structural mismatch: a networked threat met with fragmented defences, often constrained by technology limitations and regulatory controls around personally identifiable information, data sovereignty and tipping-off risks.
Criminal networks, by contrast, are adaptive and collaborative, sharing information freely and exploiting weaknesses wherever they appear. The industry is only as strong as its weakest link, with a single vulnerable point in the payment chain enough to move illicit funds across multiple institutions undetected.
Progress is emerging. Initiatives backed by the UK’s Financial Conduct Authority (FCA) are showing how firms can collaborate safely. Napier AI’s work with the FCA, The Alan Turing Institute and Plenitude on synthetic data created fully synthetic datasets built from anonymised transaction patterns, letting institutions train and refine detection strategies without exposing sensitive information.
The next phase extends beyond shared datasets into real-time intelligence sharing, where risk signals move securely across a network rather than staying trapped inside institutional walls. Achieving this will demand coordinated regulatory frameworks, trusted data-sharing mechanisms and potentially national or industry-wide utilities, alongside a mindset shift towards treating well-governed data as a collective asset.
Institutions need not wait, however. In a recent project within the FCA’s Supercharged Sandbox, Bastiman modelled transactions as a flowing system, akin to water in a river. Illicit funds create subtle ripples that propagate downstream, and by analysing their frequency and amplitude, firms can detect anomalies far from the original injection point, even without full network visibility.
On governance, Napier AI stresses that regulators are focused on outcomes rather than the intricacies of AI models. Firms adopting AI-driven detection must consistently demonstrate accuracy, explainability and auditability, which Bastiman describes as foundational rather than optional. Done correctly, AI does not introduce opacity into compliance, it removes it.
For more insights, read the full report here.
Copyright © 2026 FinTech Global
