Green AI: Sustainability-Focused Artificial Intelligence Approaches and Future Strategies

Green artificial intelligence (AI), in conjunction with sustainability, is an emerging interdisciplinary field. Its primary focus is the enhancement of the environmental sustainability of AI technologies, in addition to the leveraging of AI to address ecological challenges. Since its inception around 2019, Green AI has garnered attention due to the escalating concerns about the energy consumption and carbon footprint associated with advanced AI systems, particularly large language models (LLMs). This movement advocates for responsible technological development that aligns with global sustainability goals and addresses the pressing ecological impacts of AI advancements.Green AI encompasses two primary variants: The concept of "Green-in-AI" aims to minimise the environmental impact of AI technologies themselves by optimising algorithms and energy use. In contrast, "Green-by-AI" employs AI to address environmental issues, such as enhancing agricultural practices and biodiversity monitoring. As organisations increasingly rely on AI for various applications, the potential for significant reductions in greenhouse gas emissions through efficient energy management and resource allocation has become a focal point of research and application (Yang J., E.i. 2019; Brevini, B. 2020). Notwithstanding the promise of Green AI, notable controversies persist regarding its dual role as both a tool for sustainability and a contributor to environmental degradation. To illustrate, the energy-intensive processes necessary to train voluminous AI models can engender considerable carbon emissions, thus giving rise to questions surrounding the overall ecological merits of such technologies. Projections indicate that data centres, which are pivotal for AI operations, may consume up to 21% of global energy demand by 2030. This necessitates an urgent reevaluation of energy practices within the tech sector. As the field evolves, it faces challenges such as algorithmic bias, regulatory inadequacies, and the need for enhanced energy efficiency. The term "stakeholders" is employed to denote the individuals or groups with a vested interest in the subject under discussion. Ultimately, the future of Green AI will depend on balancing technological innovation with sustainable practices to foster a greener and more equitable world.