Artificial intelligence (AI) holds the potential to be a beacon of hope for sustainability or a harbinger of destruction. If not guided properly, the resources (e.g., water, land, and materials) and emissions (e.g., greenhouse gases) from the rapid expansion of AI could overshadow the potential AI applications for resource and emissions optimization. Moreover, sustainable AI is one of the core principles of human-centered AI (HCAI). The objectives and scope of this chapter cover both the environmental burdens of enabling AI—such as through manufacturing servers and energy consumption—and the opportunities of AI to positively contribute to environmental sustainability. The chapter begins by presenting considerations for the environmental sustainability impacts of AI, such as how data centers contribute to emissions and resource consumption. It continues with the identification of opportunities where AI can improve environmental sustainability, evaluates these opportunities and considerations with case studies, and provides recommendations for guiding sustainable AI into the future (Tran, Semiconductor manufacturing energy consumption: How green is the chip industry? PatentPC. (July 17, 2025). Retrieved from https://patentpc.com/blog/semiconductor-manufacturing-energy-consumption-how-green-is-the-chip-industry-latest-stats, 2025). At its core, Earth will survive without humans, but humans will not survive without a habitable planet. Ensuring that our technological advances do not compromise the habitability of the Earth is at the center of all human endeavors—including AI.