<![CDATA[The integration of artificial intelligence (AI) and remote sensing (RS) has transformed Earth observation (EO) by enabling the automated, efficient, and precise analysis of large and complex datasets. Rapid advances in machine learning (ML) and deep learning (DL) have further enhanced the processing and interpretation of RS data. This review examines the progress, persistent challenges, and future prospects of integrating AI with RS. Relevant literature was identified through searches of Scopus, Web of Science, IEEE Xplore, and Google Scholar using combinations of terms related to AI, ML, DL, and RS. The reviewed literature indicates that AI substantially improves the efficiency and precision of RS data processing and interpretation while expanding opportunities for automation and data-driven decision-making. Nevertheless, the broader adoption of AI-driven RS remains constrained by data quality and heterogeneity, computational demands, limited model generalisability and explainability, and ethical concerns. Future research should prioritise model efficiency, interpretability, and adaptability, alongside multimodal learning, unsupervised and semi-supervised learning, and real-time AI deployment for global-scale applications. This review consolidates current advances and limitations in AI-enabled RS and provides a research agenda for developing more efficient, interpretable, and flexible models for Earth observation applications.]]>
Copyrights © 2026
