<p>Satellite communication plays a critical role in enabling global coverage, real-time data transmission, and remote environmental monitoring. Despite its advantages, the system faces challenges related to scalability, energy efficiency, spectrum usage, and security, especially in complex space environments. This review explores the role of artificial intelligence (AI) in addressing limitations within satellite communication systems. The analysis emphasizes the integration of machine learning and deep learning models to enhance adaptability, resource optimization, and threat mitigation. Particular focus is placed on the Sentinel satellite constellation and its alignment with sustainable development goal 6(SDG6) (clean water and sanitation). A detailed examination of existing satellite communication frameworks is conducted, alongside an evaluation of AI techniques applied to space operations. AI-driven models for land classification, satellite image interpretation, object detection, resource scheduling, and collision avoidance are reviewed, which use Sentinel satellite data as a key reference for real-world applications. AI-enhanced systems outperform conventional satellite frameworks in monitoring accuracy, anomaly detection, dynamic bandwidth allocation, and environmental analysis. The integration of real-time AI monitoring and high-resolution imagery supports early detection of intrusions and natural resource tracking. Identified limitations include dependence on dataset quality, algorithmic complexity, and vulnerability to cyber threats. Integrating AI into satellite communication improves decision-making, scalability, and operational resilience. This review highlights the transformative impact of AI across various space applications while encouraging further advancements in data quality, algorithm robustness, and secure deployment strategies in mission-critical systems.</p>