Artificial intelligence for climate-smart agriculture: Enhancing food security and plant adaptation

Global climate change is an accelerating, multifaceted threat to food security and agricultural stability, requiring innovative solutions that surpass the efficacy of conventional breeding and farming practices. This review synthesizes recent advances in AI-driven approaches for climate-smart agriculture, emphasizing their unique and transformative potential in accelerating climate adaptation, optimizing resource use. We examine AI's multifaceted applications across four critical domains: high-throughput precision agriculture, accelerated genetic engineering, advanced crop yield modeling, and granular climate and pest forecasting. Specifically, we detail how AI-driven tools-including IoT sensor networks, computer vision models for phenotype screening, and deep learning algorithms-enable real-time, plant-specific nutrient and water management. Furthermore, the review illustrates how AI has the potential to markedly support and accelerate the discovery and validation of stress-resilience genes. Critically, we address the significant ethical and structural challenges impeding AI adoption, including data heterogeneity and scarcity, the potential for algorithmic bias to widen existing resource gaps, and barriers to equitable access for smallholder farmers. A key achievement is the synthesis of AI's utility in predicting crop performance under future environmental scenarios and providing actionable, site-specific recommendations to farmers and policymakers. We conclude by advocating for essential policy and governance pathways, emphasizing the necessity of transparent international data-sharing frameworks and inclusive technology transfer to ensure that AI's benefits are harnessed effectively and equitably, thus strengthening global agricultural resilience against future climate shocks.