China is leveraging artificial intelligence (AI) to transform agriculture, particularly in smart breeding, which is paving the way for high-yield, climate-resilient "super crops" that could reshape global food security amid environmental challenges.
At the 2025 Seed Congress, held from March 20 to 23 in Sanya, Hainan Province, smart breeding was a focal point. More than 40 exhibitors presented state-of-the-art technologies and flagship crop varieties, demonstrating how AI-driven breeding is transitioning from concept to reality. Digital exhibits showcased breakthroughs such as AI-powered genomic analysis and cloud-based farm monitoring systems.
Experts at the event highlighted how smart breeding is modernizing traditional techniques into precision agriculture. This transformation is helping to address seed resource constraints, strengthen self-sufficiency in the seed sector, and bolster China's strategic role in the global seed market.
Li Jiayang, an academician at the Chinese Academy of Sciences (CAS), introduced the idea of "intelligent creation of intelligent varieties," emphasizing the integration of AI, biotechnology, and information technology to develop crops that can adapt to environmental conditions.
Another CAS academician, Qian Qian, noted the complexity of crop traits. "Understanding the relationship between genes and traits requires computational power and advanced algorithms," he said. Unlike conventional small-scale breeding, smart breeding follows an industrialized model, pooling extensive resources to efficiently integrate desirable traits.
AI-driven breeding is already yielding significant results. At China National Seed Group, researchers employ a cloud-based system to remotely monitor fields, collecting real-time data on crop health and growth. This allows scientists to address issues promptly and optimize breeding strategies.
One major advancement is in hybrid rice development. Traditionally, breeders tested thousands of combinations to identify a superior hybrid. Now, AI-powered genomic analysis can predict high-yield combinations before field trials commence.
Chinese researchers use Earth science satellite to monitor offshore oil, gas platforms
"We have moved from experience-driven to data-driven breeding," said Li Huihui, deputy director of the National Nanfan Research Institute of the Chinese Academy of Agricultural Sciences.
Another key innovation is "Fengdeng," a seed-focused large language model developed primarily by the Yazhou Bay National Laboratory in Sanya. This AI platform provides expert guidance on breeding, cultivation, and industry trends, equipping farmers and researchers with valuable insights.
Despite these advancements, challenges persist. China's smart breeding sector lags behind global seed industry leaders in data-sharing infrastructure and commercialization. The fragmentation of datasets limits AI’s ability to predict and engineer superior crop varieties.
"Accelerating the development of high-yield, high-quality and climate-resilient 'super varieties' is crucial," Qian said, calling for collaboration among breeding institutions, AI researchers, and agribusinesses to propel smart breeding innovation.