Napier AI has responded to a consultation from HM Treasury, the UK government’s economic and finance ministry, on enhancing the effectiveness of the Money Laundering, Terrorist Financing, and Transfer of Funds (Information on the Payer) Regulations 2017 (MLRs). These regulations require businesses to identify and prevent money laundering and terrorist financing.
Napier AI appreciates HM Treasury’s initiative to update the guidance on Customer Due Diligence (CDD) regulations and improve the coordination of systems involved in combating financial crime. However, they caution that making such information public could potentially aid criminals in understanding and circumventing these defences.
The importance of CDD in the initial risk assessment and onboarding of new customers is well recognised. Nonetheless, Napier AI points out that Enhanced Due Diligence (EDD) often relies on binary rules that fail to capture the complexity of modern financial crime. They advocate for a Risk-Based Approach (RBA) in CDD, which should be multi-faceted and trigger responses from various sources, rather than being limited to onboarding. Napier AI argues for a continuous, smart, and multi-faceted approach to client risk assessment, aiming for Perpetual Client Risk Assessment (pCRA) processes to build a robust defence against financial crime.
In their response to HM Treasury, Napier AI emphasised two main points: the necessity for smarter, multi-faceted, and dynamic triggers instead of fixed thresholds, and the inclusion of more factors and data points in financial crime risk calculations, moving towards pCRA.
Napier AI criticises fixed thresholds, as they can allow criminals to manipulate their transactions below these limits to avoid CDD checks. They suggest using smart, multi-faceted techniques like dynamic segmentation, which can identify when a customer’s transactional behaviour deviates from their segment norm. This approach makes it more challenging for criminals to evade detection by focusing on overall behaviour rather than individual thresholds.
For instance, if a student begins exhibiting transactional behaviour similar to a crypto trader, dynamic segmentation can trigger CDD checks. This method leverages multiple AI models to consider numerous factors, detecting changes in material risk associated with the customer’s profile, even if individual changes seem minor.
Additionally, Napier AI explains that during societal shifts or external events, inflexible thresholds can generate excessive false positives, necessitating manual system adjustments. They cite the example of increased online payments by small restaurants during the COVID-19 pandemic, which was not unusual given the circumstances and should not trigger unnecessary alerts.
Napier AI also addresses the need for continuous risk assessment beyond the onboarding stage. They note that criminals often conduct money laundering soon after opening an account. Traditional CDD/EDD processes, focused on onboarding and periodic reviews, create a lag that criminals exploit. Continuous assessment, integrating numerous data points, can constantly recalibrate customer risk and quickly identify material risks.
For example, if a new account exhibits unusual activity soon after onboarding, even within acceptable thresholds, a pCRA system can launch immediate CDD/EDD, flagging potential AML typologies. Napier AI believes that pCRA represents the future of financial crime compliance, offering smarter detection of risky activities and reducing unwanted noise, allowing analysts to focus on genuinely suspicious behaviour.
Read the full story here.
Keep up with all the latest FinTech news here
Copyright © 2024 FinTech Global
