The increasing adoption of artificial intelligence (AI) in healthcare and other industries has resulted in a surge of data generation, model complexity, and computing requirements. While AI holds the potential to revolutionize diagnostics and decision-making, its environmental impact must not be overlooked. This chapter addresses the urgent need for energy-efficient AI models and sustainable data management practices, framing the conversation within the emerging “third wave” of AI ethics that centers on environmental responsibility. We explore the carbon footprint of model training, present metrics and tools for emission tracking, and offer guidance on how to minimize energy consumption through model optimization strategies such as pruning, quantization, and distillation. In parallel, we emphasize the role of responsible data storage, including the management of dark data and cold archiving practices, as essential to lowering the environmental toll of digital infrastructures. With practical examples, this chapter advocates for embedding sustainability at the core of AI development and deployment. It calls on the global AI community of researchers, developers, industry leaders, and policymakers to adopt energy-aware approaches that ensure long-term ecological balance without sacrificing technological progress.