Macao Financial Institutions Face Challenges in Sanctions Screening

Introduction

A recent industry roundtable hosted by Capgemini, in partnership with Hummingbird and WorkFusion, brought together senior representatives from various financial institutions to discuss the challenges they are facing in screening processes. The key issues related to existing screening processes include:

• High false hit rates
• Complexity of Russia sanctions
• Lack of information on counterparties
• Budgeting and staffing constraints
• “Noise” or insufficient data surrounding alerts

Operational Challenges in Trade Finance

The roundtable also highlighted the operational challenges in the trade finance space, including manual inputs from letters of credit and other documents resulting in numerous errors.

Exploring AI-based Solutions

AI-based tools are being explored to address these challenges, with current applications primarily on the alert adjudication side. However, there is a need for AI intervention in up-front screening as well.

Implementing Modern AI-based Tools

Implementing modern AI-based tools presents challenges that require careful planning, internal support, collaboration between IT and business units, attention to regulatory imperatives, and a strategic approach to ensure a smooth integration.

Some organizations have replaced first-level human review of screening alert hits with purpose-built algorithms for screening names and payments and analyzing them in real-time.

The Process of Incorporating AI into Screening Systems

The process of incorporating AI into a screening system includes:

• Model selection, training, testing, and validation
• Integration of models with existing systems
• User training and adoption
• Ensuring continued compliance with all applicable laws and regulations

Implementation Challenges

Implementation challenges identified by the roundtable participants include:

• Model governance
• Skills gaps
• Legacy system compatibility
• Data security concerns
• User training and adoption

Regulatory Acceptance of AI-based Tools

Given the highly regulated nature of the financial industry, regulatory acceptance is a key consideration in implementing AI-based tools. Ensuring explainability, transparency, accountability, proper model risk management, and data management are important aspects to regulatory acceptance.

Conclusion

In sum, the benefits of employing modern AI-based tools to enhance screening processes are compelling, but implementing these tools presents challenges that require careful planning, internal support, collaboration between IT and business units, attention to regulatory imperatives, and a strategic approach to ensure a smooth integration.