Auditable Climate Risk Intelligence from Fragmented ESG Data: Deterministic Orchestration and Imbalance-Aware Learning for Scope 1-3 Validation
Researchers have developed a new framework for validating ESG and climate risk data, addressing the fragmentation and lack of auditability in current systems. The proposed method integrates a single source of truth orchestration, temporal anomaly detection, and ensemble learning with a focus on explainability and governance. To facilitate open reproducibility, a synthetic ESG validation benchmark has been created and released, calibrated against established standards like the GHG Protocol and ISSB. AI
IMPACT Introduces a novel AI-driven approach to improve the accuracy and auditability of climate risk reporting.