This study introduces AiBEDO, an AI model leveraging the Fluctuation-Dissipation Theorem (FDT) to analyze climate intervention scenarios. AiBEDO is trained on a large Earth System Model (ESM) dataset to predict climate responses to external forcings, significantly accelerating the evaluation of interventions like Marine Cloud Brightening (MCB). The model effectively captures complex, multi-timescale effects, enabling rapid prototyping and optimization of MCB strategies to achieve regional climate targets and prevent tipping points. The methodology is generally applicable to other computationally intensive scientific domains.

Soo Kyung Kim, Kalai Ramea, Rühling Salva, Haruki Hirasawa, Subhashis Hazarika, Dipti Hingmire, Peetak Mitra, Philip J Rasch, Hansi A Singh