This paper proposes an innovative approach to enhancing energy efficiency in Optical Wireless Communication (OWC) networks by integrating Artificial Intelligence (AI)-driven green management strategies. The increasing demand for sustainable communication technologies is determined by a combination of environmental, regulatory, economic, technological, and social factors. As the global digital infrastructure continues to expand, the adoption of sustainable practices in communication technologies is essential for mitigating environmental impacts and ensuring long-term viability. By leveraging machine learning algorithms and advanced data analytics, the study aims to develop sustainable and efficient management practices of the OWC networks with the aim to reduce the environmental impact while maintaining high performance and reliability of the communication link. This paper explores the various ways AI is applied to optimize OWC networks, highlighting key areas and real-world applications. We propose the use of a Machine Learning (ML)-based Random Forest algorithm to implement a soft-switching strategy in urban OWC/RF networks with the aim to enhance the optical network efficiency by dynamically adjustment of transmission parameters or switching to RF depending on the real-time conditions.