Artificial Intelligence (AI) plays a dual role in the pursuit of sustainable development, serving both as a tool for advancing sustainability initiatives and as a technology requiring sustainable governance. Current research examines the paradigm adjustments necessary to include ethical AI into sustainable development initiatives. This paper presents an up-to-date Systematic Literature Review (SLR) of 35 studies using PRISMA methodology, categorizing research into two key themes: AI for Sustainability and Sustainability of AI. The first theme explores how AI-driven innovations contribute to individual, social, economic, environmental, and technical sustainability. The second theme examines the environmental and ethical implications of AI itself, focusing on parameters like ethics, fairness, explainability, accountability, privacy, and security. Our findings highlight both the transformative potential of AI in supporting sustainability and the urgent need for ethical frameworks, regulatory oversight, and interdisciplinary collaboration to ensure AI's sustainability. The study concludes by proposing the paradigm shifts that bridges AI ethics and sustainability, promoting responsible AI adoption in alignment with long-term global sustainability goals. The paradigm shifts explored in this research paper hold profound practical implications across industries, research endeavors, and policymaking by offering a roadmap for the responsible use of AI in a manner that contributes meaningfully to sustainable development initiatives.