Geopolitical Risk, Ai, and Investment Finance: How Do They Intersect?

I’m about to embark on a trip to Montreal soon to meet some members of the AI team and collaborate with PRS’ marketing cohorts. Visiting the city is always a delightful experience, and it has truly become the premier global hub for AI, driven by pioneering research and a thriving ‘ecosystem’ of corporate labs and startups.

As we approach 2026, I’ve been frequently asked by clients and others about the evolving nature of ICRG data and its applications across various sectors, particularly finance. In response, I’ve compiled this note as an update.

Unlike many AI models that rely on potentially biased open-source data, ICRG data undergoes rigorous vetting and normalization through multiple avenues. This ensures the accuracy, verification, and competitive advantage of AI applications built using this data over those utilizing publicly available platforms.

The dataset employs a quant-driven model covering 141 countries, transforming subjective political analysis and objective financial data into precise “risk points.” This structured format is ideal for integration into machine learning models, enabling accurate market return prediction and the identification of future risk events.

ICRG maintains the longest-running database of political risk data, spanning back to 1984. This consistency allows AI to train on decades of historical patterns, enhancing long-term forecasting capabilities.

The ICRG methodology distinguishes political, economic, and financial risk variables, helping AI developers avoid the common error of “double counting” systematic risk.

Notably, ICRG data and methodology are the only ones accepted by courts for valuing political risk in commercial disputes. They also serve as the primary data source for Transparency International’s Corruption Perceptions Index.

The ICRG model offers users the flexibility to modify the weighting of its 22 variables to suit specific project requirements, catering to niche AI risk assessments. However, it’s essential to double-check responses, as AI can make mistakes.

In terms of finance and AI, our ICRG data is utilized to enhance predictive analytics, improve risk management, and optimize investment strategies. It provides a high-quality, quant-driven dataset on political, economic, and financial risks for over 140 countries.

For instance, AI models integrate ICRG’s risk ratings and metrics, which have been proven predictive of equity market returns, to predict marketing returns. This information enables traders and portfolio managers to optimize asset allocation and country exposure.

AI systems leverage the data’s ability to predict future risk events for risk management purposes. Machine learning models trained on the historical data, spanning back to 1984, identify long-term patterns and improve long-term forecasting of political risk events.

In credit risk management, AI algorithms utilize the quantifiable risk data to perform more accurate credit scoring and default prediction for individuals or entities with exposure to international operations. This integration with other data sources enhances the accuracy of credit risk assessments.

Natural Language Processing (NLP) tools combined with ICRG data help compliance officers stay informed about evolving international regulations and political risks. This automation streamlines monitoring and reporting processes.

AI platforms can process both structured, numerical ICRG data and unstructured data (such as news feeds) to build a comprehensive view of potential risks. This near-real-time capability allows for dynamic adjustment of risk scores as new information becomes available, providing a significant advantage over traditional, static risk assessment methods. This integration enables financial institutions to enhance the accuracy of their risk assessments and make faster, more informed decisions.

Our Data Drives.

The PRS Group

CHRISTOPHER MCKEE, PHD CHIEF EXECUTIVE

Christopher McKee is PRS’ CEO and Owner. An international political economist, global investor, entrepreneur, and author, Chris received his PhD from Queen’s University (Canada) and has been involved in the field of geopolitical risk, limited recourse financing, and private sector development for the past 25 years.