Researchers presented the TAISE framework for modeling catastrophe risks

A group of scientists has developed the TAISE framework, designed for automated creation of extreme weather scenario scripts. The system uses modern AI weather forecasting models to generate continuous global atmospheric fields, replacing traditional manual modeling methods that have remained largely unchanged since the 1990s.

Experimental verification has shown that using TAISE allows for tenfold reduction in computational costs compared to classical approaches. Unlike standard methods based on data "snapshots", the new framework provides high temporal coherence and takes into account interregional correlations, which is critical for accurate risk assessment.

The technology works through a mechanism of self-iterative generation, allowing for the modeling of complex climate events with high detail. This solution is aimed at democratizing the process of catastrophe risk assessment, making it accessible to insurance companies, reinsurers, and government agencies involved in risk management.

The implementation of such tools allows for a transition from static models to dynamic and comprehensive portfolio evaluation. The source code and research methodology are available under the Non-exclusive distribution license, opening up opportunities for the integration of AI approaches into the financial sector and disaster protection systems.