Green Artificial intelligence Foundations, Applications, and Pathways to Sustainable Development

The fast evolution of artificial intelligence (AI) systems has worried people about their environmental impact thus prompting the rise of Green AI. In the present systematic review, we are going through the 32 articles published in peer-reviewed journals that were analyzed based on PRISMA standards regarding the conceptual bases, applications, and the future of Green AI. The review identified three paradigms: Green AI (computational efficiency), Sustainable AI (holistic socio-technical responsibility), and AI for Green (AI applied to sustainability challenges). A large part of the resources that would be used for the environments, monitoring, agriculture, and smart city applications can be saved by 15-30% through Green AI. The main difficulties are performance and efficiency balancing, limiting budget, and a research mentality that values precision more than sustainability. The research points out the dual function of AI in environmental matters as that of polluter and of a device for making the planet greener through humane practices and technologies. To sustainable AI, efficient algorithm design, regulatory support, the establishment of carbon-aware metrics, and collaboration among different disciplines to create the adoption of AI that is both economical and ethical are needed