Preventing and Combating Terrorism Financing Using AI

Background

Combating terrorism financing has long been a global challenge. To address this, the United Nations Office of Counter-Terrorism launched the goFintel project, an initiative aligned with UN Security Council Resolutions 2462 and 2482. goFintel is designed as a next-generation platform to detect, prevent, and counter both money laundering and the financing of terrorism. This project enhances Member States’ capabilities through secure data integration, advanced analytics, and cross-functional intelligence sharing across financial, judicial, and law enforcement domains.

Objective

The primary goal of this project  was to build a robust AI model capable of:

• Correlating financial, criminal, and other data sources to identify terrorism financing patterns.
• Visualizing actionable insights in real-time to assist Financial Intelligence Units (FIUs) and other stakeholders.
• Empowering Member States to improve operational capabilities in detecting and preventing financial crimes.

Approach

To tackle the problem, Omdena team of 50 global AI changemakers collaborated with the UN’s goFintel project over two months. The project followed an open-source methodology and employed advanced machine learning techniques. Key steps included:

1. Data Collection & Preparation:
   • Leveraged financial crime datasets, including:
     • Suspicious Transaction Reports (STRs) and Cash Transaction Reports (CTRs).
     • United Nations Security Council Consolidated List.
     • Open-source intelligence (OSINT), company registers, and social media data.
2. Model Development:
   • Utilized supervised and unsupervised learning methods to establish correlations across diverse datasets.
   • Designed algorithms to analyze patterns based on names, times, locations, transaction amounts, and travel records.
3. Visualization & Real-Time Insights:
   • Built dashboards for dynamic, real-time visualizations to display data insights.
   • Delivered actionable outputs accessible to stakeholders for decision-making.

Results and Impact

The collaboration resulted in the successful engineering of a comprehensive AI model that:

• Correlated data effectively across financial, criminal, and other records, providing unprecedented insights into terrorism financing patterns.
• Enabled real-time visualizations to support intelligence operations and decision-making processes.
• Improved operational efficiency for Financial Intelligence Units (FIUs) by expanding their capabilities to analyze multiple data sources securely and cohesively.

This project strengthens the global fight against terrorism financing, equipping Member States with innovative tools to prevent financial crimes and support counter-terrorism operations.

Future Implications

The findings and methodologies developed through this project have far-reaching implications:

• Policy Development: Governments and international bodies can leverage insights to create more effective anti-money laundering (AML) and counter-terrorism financing (CTF) policies.
• Enhanced Collaboration: The secure, multi-stakeholder framework provides a blueprint for future collaborations between financial institutions, law enforcement, and judicial entities.
• Advancing AI in Security: The open-source nature of this project paves the way for further research and innovation in using AI to combat financial crimes and terrorism financing.

This initiative marks a significant step forward in leveraging technology to address global security challenges, laying the groundwork for scalable and sustainable solutions.