Toward Transparent Carbon Trading: Integrating Explainable AI and GANs for Circular Economy-Driven Price Prediction

AAAI 2026 Workshop AIGOV Submission13 Authors

Keywords: AI, Disruptive Technology, XAI, GANs, Carbon Price Prediction, Carbon Trading, Sustainability, SDGs, Carbon Emissions, Net Zero Emissions.

TL;DR: Development of an GAN-XAI based prediction mechanism for improving carbon price interpretability and forecasting for a resilient CE lifecycle.

Abstract: The need to actively involve businesses in safeguarding the environment has grown exponentially. Whilst initiatives like the Emissions Trading Scheme (ETS) and Circular Economy (CE) have emerged as key to propagate the same, an opportunity to unite them emerges due to the imminent benefits observed, like supply chain optimization, better emission management, etc. However, from a real-life standpoint, integration of both emerges to be difficult. This is owing to issues like the inability of traditional ETS systems to consider complex parameters, current AI-driven trading mechanisms failing to accommodate SDGs and other environment-centric governing factors for price prediction, a lack of CE-centric AI/ML-driven solutions on an organizational level, etc. Hence, to overcome the gaps identified, a hybrid XAI-GAN-driven framework for interpreting carbon prices for resilient CE lifecycle development is proposed for development. Preliminary findings on our established framework display a strong inclination towards seamless integration of ETS and CE practices from an emerging market viewpoint and heavily advocate for its potential to provide tangible insights for concrete emission mitigation from a real-world perspective.

Submission Number: 13