Agricultural Sciences: Techniques and Innovations Vol. 11

Artificial intelligence is increasingly being integrated into agricultural systems to support environmental sustainability, resource efficiency, and climate-resilient production. This review evaluates artificial intelligence-based technologies in crop and livestock production and presents a SWOT-based assessment of their integration into agriculture. A narrative and thematic approach was followed, drawing on peer-reviewed studies that directly address artificial intelligence and environmental sustainability in agriculture. The evidence was synthesised across crop production, livestock production, general environmental sustainability, and SWOT analysis dimensions. In crop production, artificial intelligence supports precision irrigation, fertilisation management, water-stress monitoring, disease detection, weed monitoring, targeted spraying, and resource-use optimisation. The reviewed evidence indicates water savings of 27.6-40% in tomato irrigation, 35.1% in a paddy and vegetable test plot, 29% in maize irrigation, and up to 70% in AI-powered solar irrigation systems. Artificial intelligence-based weed-control systems reduced herbicide or pesticide use by 10-70%, depending on crop, model, and system design. In livestock production, artificial intelligence supports feed-use prediction, feed-waste reduction, heat-stress recognition, methane-emission prediction, greenhouse-gas assessment, carbon-footprint estimation, emission-intensity reduction, and manure-management decisions. The SWOT analysis identifies resource efficiency, environmental monitoring, evidence-based decision support, and climate-resilient production management as the main strengths of artificial intelligence. The main weaknesses are data quality, model generalisation, infrastructure cost, interoperability, digital literacy, privacy, and farm-level validation. Opportunities include edge-cloud systems, low-cost IoT solutions, solar-powered irrigation, remote sensing, farmer-centred design, and AI-based emission monitoring. Threats include climate variability, unequal access to technology, data-governance risks, and environmental costs linked to the life cycle of digital technologies.