GREEN ARTIFICIAL INTELLIGENCE: DESIGNING ENERGY-EFFICIENT MACHINE LEARNING MODELS FOR SUSTAINABLE COMPUTING

The rapid expansion of Artificial Intelligence (AI) has significantly improved automation, data analytics, and intelligent decision-making across numerous sectors. However, the increasing complexity of modern AI models has resulted in substantial computational requirements and energy consumption, raising concerns regarding environmental sustainability. Green Artificial Intelligence (Green AI) has emerged as an important research field that emphasizes the development of energy-efficient algorithms, resource-aware computing, and environmentally responsible AI systems. This paper presents a comprehensive review of Green AI, examining its principles, enabling technologies, applications, challenges, and future research directions. The study explores optimization techniques including model compression, pruning, quantization, knowledge distillation, and efficient neural network architectures. Additionally, it discusses the integration of renewable energy resources, cloud optimization, edge intelligence, and carbon-aware computing for sustainable AI deployment. The findings demonstrate that Green AI can substantially reduce computational costs and carbon emissions while maintaining acceptable predictive performance, thereby contributing to environmentally sustainable digital transformation.