A familiar predicament haunts the global payments landscape: can financial institutions outgrow their compliance issues without stretching resources to the breaking point? Scaling staff often means soaring costs and growing complexity. RegTech company Hawk AI has outlined how it helped an payment processor improve their operations.
For Company A, an international payment processor dealing with over one billion transactions across numerous markets annually, the answer was embedded in AI. Using Hawk AI’s explainable technology, the firm has enhanced its anomaly detection, unearthed suspicious activities and filed SARs with more efficiency.
Company A sits at the crossroads of commerce, processing payments in 37 countries for 400 merchants using 700 methods. Hawk AI has brought substantial benefits, most notably False Positive Reduction and Auto-Closing. The results are striking: a 97.3% reduction in false positives, a 51.4% true positive rate, reasonable thresholds, improved case relevance, manageable caseloads, explainable auto-closing, a risk-based approach, and quality control.
But even with Hawk AI’s anomaly detection at work, Company A’s small compliance team struggled under an enormous caseload. They adopted high case thresholds to cope with the workload, but this approach opened them to regulatory scrutiny. Capacity-based thresholds limited the range of cases and merchants they could investigate and created regulatory “black holes.” Further, their investigative team was swamped by false positives, raising the risk of missing true positives and falling foul of regulatory requirements.
To alleviate these issues, Company A leveraged Hawk AI’s False Positive Reduction and Auto-Closing. This solution assesses each rule hit within the case context, determining whether to close the case automatically or escalate it to an analyst. Hawk AI’s Wolfgang Berner explained, “Using a context-aware AI for filtering is much more effective. It can create better decision boundaries and isolate suspicious activity from false positives.”
With Auto-Closing, Company A could set thresholds based on actual risk, not staff numbers. The thresholds were set to the team’s standards of suspicious activity, allowing the False Positive Reduction model to automatically close cases that didn’t meet all the criteria. This enabled the compliance team to review more quality cases and catch more true suspicious activity.
The relevance of cases was also significantly improved, and caseloads became more manageable. The model’s ability to auto-close false positives increased the team’s efficiency, preventing the need for hiring additional staff. By focusing on the right cases instead of more cases, the company successfully enhanced its regulatory compliance and suspicious activity detection.
Auto-Closing also proved beneficial during audits. The explainable AI left a detailed trail of natural language explanations each time a case was auto-closed. As a result, the compliance team could justify their decisions during reviews or audits, with the necessary contextual information readily available.
Hawk AI’s solution complemented Company A’s risk-based approach by allowing them to focus resources on high-risk areas. The team gained control over the types of cases they reviewed and could adjust thresholds based on the risk factors of a specific industry or segment.
Hawk AI also allowed the team to randomly sample cases closed by the model and evaluate its effectiveness, ensuring a “trust but verify” approach. With a detailed explanation of the decision to close cases available, the team could make necessary adjustments, thereby continually improving the model’s efficacy.
Read the story here.
Keep up with all the latest FinTech news here
Copyright © 2023 FinTech Global
Copyright © 2018 RegTech Analyst
