Are Climate Risk Risk Regulations Evidence-Based? Leveraging Artificial Intelligence to Bridge the Gap between Regulatory Practice and Academic Research

This article presents an analysis of the connection between emerging climate risk policies in the banking sector and academic literature. The research focuses on regulations implemented after pivotal events such as the Paris Agreement led to increased scrutiny and mandated climate risk disclosures by financial institutions. By applying artificial intelligence and natural language processing techniques, the study analyzes 16 major global climate risk regulations, categorized into three distinct categories. We compare their outcomes with 83 peer-reviewed academic articles published from 2017 to 2024 to determine whether academic research outcomes influence those regulations. The findings reveal a moderate alignment (74.2% cosine similarity) between regulatory policies and academic research, indicating that although many regulatory bodies incorporate research insights, significant gaps remain. The article provides insights into enhancing emerging climate risk policies, suggesting a need for an evidence-based approach that integrates climate risk disclosure standards, climate risk management, and climate research into framework development. Theoretically, this study advocates for institutional collaboration among regulators through a polycentric governance system, which will help facilitate the effective integration of scientific research to strengthen climate risk policy management and development into the future.