Identifying Important AI Sub-Pillars on Carbon Dioxide Emission Using Data Mining Techniques

This paper discusses the urgent need to analyze the environmental effects of artificial intelligence (AI) against its rapid growth worldwide. We explore how AI components and economic growth affect carbon emissions (CO2) in 52 countries between 2019 and 2022. In this paper, we analyzed seven AI sub-pillars: commercial, development, government strategies, infrastructure, operating environment, research, and talent, alongside economic growth by using machine learning algorithms. Random Forest demonstrates robust accuracy across diverse datasets. Moreover, instead of utilizing all the pillars of AI, only commercial, development, government strategies, infrastructure, research and economic growth attributes provide the highest accuracy level. The novelty of this research is the use of machine learning methods to provide economic implications and offer policymakers and industry leaders actionable insights to align AI development with global sustainability goals.