Machine Learning and Artificial Intelligence for Data-Driven Sustainable Agriculture: Applications, Challenges, and Opportunities.

Abstract The integration of machine learning (ML) and artificial intelligence (AI) is rapidly transforming sustainable agriculture by enabling data-driven decision-making, enhancing productivity, and supporting resource management. This review evaluates recent advances in AI and ML applications across core domains such as crop disease detection, soil health assessment, yield prediction, crop phenotyping, resource optimization, and supply chain management. Methodologies including remote sensing, VIS-NIR spectroscopy, deep reinforcement learning, and IoT-enabled systems are examined for their contributions to real-time monitoring and enhanced decision support. Despite clear benefits, widespread adoption faces barriers such as data quality, model transparency, scalability, and technological accessibility. The review discusses opportunities for integrating AI/ML with emerging technologies (UAVs, big data, cloud computing), fostering user-centric tools, and advancing policy frameworks to drive sustainable development. The findings highlight the importance of continued interdisciplinary research to address outstanding challenges and harness AI-driven innovation across diverse agricultural settings.