Advancing climate-resilient rice breeding through genomics, speed breeding, and artificial intelligence

Abstract

Global rice production is increasingly threatened by climate change, shrinking arable land, and the growing demand driven by rapid population growth. Escalating abiotic stresses and intensified biotic pressures pose major challenges to rice productivity and global food security. Conventional rice breeding approaches, although historically successful, are becoming increasingly inadequate because they rely on lengthy selection cycles, environment-dependent phenotyping and limited precision in improving complex polygenic traits. This review discusses the limitations of traditional breeding methodologies in the face of increasingly unstable climatic conditions and highlights the transition toward climate-smart breeding systems. Particular emphasis is placed on integrating advanced biotechnological approaches, including marker-assisted selection (MAS), genomic selection (GS), CRISPR/Cas-based genome editing and speed breeding. These strategies collectively enhance selection efficiency, accelerate genetic gain and shorten breeding cycles, thereby facilitating the development of stress-resilient cultivars. Furthermore, the review explores the emerging role of artificial intelligence (AI) in integrating multi-omic, phenotypic and environmental data to optimize breeding pipelines through high-throughput phenotyping, predictive analytics and precise genotype-by-environment (G × E) modeling. The discussion also addresses ideotype design frameworks as strategic tools for accelerating crop adaptation to future environmental challenges. Finally, we propose an integrated roadmap for sustainable rice improvement that combines genomics, genome editing and AI-driven decision-making. This framework offers a comprehensive strategy to bridge the adaptation gap and support the development of high-yielding, climate-resilient and nutrient-enriched rice varieties that sustain productivity under future climatic fluctuations.