The increasing adoption of Artificial Intelligence (AI) in finance raises growing concerns about its environmental footprint, particularly energy consumption and carbon emissions. Finance systems compute millions of model inferences, forecasts, and real-time decisions each day. While Green AI emphasizes energy-efficient and sustainable AI practices and has advanced rapidly in domains such as computer vision and natural language processing, its adoption in finance remains underexplored. This study presents a systematic literature review (SLR) and proposes a new approach for implementing Green AI models in finance. We analyze 58 peer-reviewed studies published between 2018 and 2025 and retrieved from the Scopus database to assess the state of Green AI in financial applications. The SLR identifies major gaps, including the absence of standardized benchmarks and assessment tools for Green AI in finance. It also highlights a persistent trade-off between reducing computational costs and maintaining high predictive accuracy. This tension complicates deployment in real-world financial settings, where economic benefits are often prioritized. Green AI can also help democratize access to advanced analytics, especially for smaller financial institutions that lack substantial computing resources. Finally, we present a taxonomy of Green AI techniques mapped to four stages of the machine learning lifecycle: data preparation, architecture design, model development, and deployment. We propose a theoretical framework that integrates Green AI principles (e.g., model pruning) with energy-monitoring tools to guide sustainable AI adoption in finance. The framework supports financial institutions and policymakers in implementing responsible AI systems that balance performance, compliance, and environmental sustainability.
Elbouknify et al. (Mon,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: