The rapid expansion of artificial intelligence (AI) technologies has significantly increased global energy consumption, raising serious concerns about environmental sustainability. This study presents a comprehensive framework for integrating green energy solutions into AI infrastructure to reduce carbon emissions from training and deploying large-scale machine learning models.
The framework uses a layered approach that incorporates renewable energy sources, energy-efficient workload management, and hardware optimisation techniques. Experimental results show energy savings of 35% to 60% while maintaining acceptable computational performance. These findings demonstrate the feasibility of sustainable AI development and provide valuable recommendations for researchers and industry professionals seeking to minimise environmental impact without hindering technological advancement.