The Bank for International Settlements has completed proof-of-concept testing of Project Gaia, an initiative conducted in concert with three central banks to explore the use of natural language processing, optical character recognition and machine learning to extract climate-related data from corporate reports.
Working with the European Central bank, Deutsche Bundesbank and the Bank of Spain, the project report describes how Gaia was able to overcome differences in definitions and disclosure frameworks across jurisdictions to make it easier to compare information on climate-related financial risks.
The proof of concept broke new ground by using large language models (LLMs) to automatically extract climate-related indicators such as total emissions, green bond issuance and voluntary net-zero commitments from publicly available corporate reports.
The project partners say the flexible design may serve as a model for AI-enabled applications in a broader range of use cases for central banks and the financial sector.
Pablo Hernández de Cos, governor of Bank of Spain, comments: "Project Gaia is an important breakthrough on the road to the understanding and adaptation of AI into the innovation culture of the Eurosystem. The potential applications and synergies of this tool are nearly limitless."