Chinese researchers are using artificial intelligence to speed up the agricultural seed breeding process, potentially revolutionizing a core sector key to national food security.
Sun Qixin, an academician of the Chinese Academy of Engineering and a professor at China Agricultural University, is using computer simulations to predict plant growth from genetic code to accelerate crop breeding by up to hundreds of times.
Previously, this would have been a labor-intensive task undertaken in the field.
Sun said the technology could mark the beginning of a fundamental revolution in agricultural breeding, with intelligent systems potentially reaching or surpassing the capabilities of individual human breeders.
For generations, crop breeding has relied heavily on the experience and painstaking work of individual breeders. They would examine plants one by one in the field, measure their characteristics, record observations and repeatedly select those with desirable traits.
Such methods made enormous contributions to agricultural development, but they are time-consuming and labor-intensive, particularly when breeders need to identify a small number of superior varieties from thousands or even millions of individual plants.
"Artificial intelligence will certainly bring a revolution in breeding in the long run," Sun said.
He added that three major foundations are needed to make that transformation possible, which are more comprehensive agricultural data, lower-cost genetic sequencing and more sophisticated models capable of connecting a plant's genetic makeup with its observable characteristics.
With drones monitoring fields from above and multispectral equipment collecting information at ground level, researchers can increasingly record the entire growth process of individual plants, from emergence to harvest. Such information can help AI systems learn how different varieties respond to environmental conditions and how their characteristics evolve throughout the growing season.
Sun said he expected that within three to five years, the accumulated technical data on major grain crops and livestock could reach a level sufficient to support the development of specialized intelligent breeding models.
"Meanwhile, the cost of genetic sequencing is reducing," Sun said, citing wheat as an example. Sequencing the genome of an individual wheat plant to a high level of quality once cost around 1 million yuan ($150,000), whereas comparable work can now be completed for about 100 yuan.
The dramatic reduction makes it increasingly feasible to obtain genetic information for large numbers of plants, providing the raw material needed to train AI models, he said.
In addition, one of the most important steps is to establish reliable connections between genetic information and physical characteristics.
Researchers need to teach AI systems to predict what a plant will look like and how it will perform based on its genes, while also identifying the genetic characteristics underlying desirable traits observed in the field.
Once such models become sufficiently accurate, they could significantly reduce the need for extensive field trials and manual selection.
Sun said AI could eventually take over as much as 80 percent of the workload involved in certain stages of conventional breeding, potentially making the process dozens or even hundreds of times more efficient.
The technology is also opening up new possibilities for understanding plant root systems, which play a critical role in crop yields, nutrient absorption and resistance to drought.
In the past, breeders often had to wait until harvest to assess whether a variety had performed well, with limited ability to observe the root structures responsible for its performance.
New imaging technologies and AI algorithms are making it possible to analyze root systems more precisely and identify traits associated with high yields, efficient fertilizer use and drought resistance.
This could allow breeders to design varieties with desirable characteristics from the outset rather than relying primarily on repeated field selection.
Sun said some specialized intelligent systems had already begun addressing specific breeding tasks, while research institutions and technology companies were exploring cooperation by combining computing chips and algorithms with agricultural expertise.
Building fully functional intelligent breeding systems would require patience and sustained investment, he said, adding that about five years are potentially needed before their practical impact becomes more widely visible.
Apart from breeding, AI, big data, satellite remote sensing, drones and autonomous machinery are changing almost every stage of agricultural production.
In an unmanned farm, for example, people no longer need to perform every operation in the field. Instead, digital systems coordinate machinery to carry out tasks ranging from planting and fertilization to irrigation and harvesting.
Such developments are helping agriculture move toward more standardized, large-scale and efficient production.
Chen Wei, president of China Agricultural University, said the changing nature of agriculture is reshaping the country's demand for talent and the development priorities of agricultural universities.
To respond to the changes, the university is adjusting its academic programs and research priorities as it develops a medium — and long-term strategy for the next five to 10 years.
The university has introduced new programs in biological breeding, food nutrition and health, green agricultural development, veterinary public health, and intelligent science and technology. It is also planning to establish schools specializing in bio-manufacturing and artificial intelligence.
"We are strengthening organized research efforts to address key technological bottlenecks in areas such as seed development, green agricultural production and agricultural machinery," Chen said.
He emphasized accelerating the commercialization of research findings. Agricultural technologies should be ultimately affordable and practical for farmers, Chen said. "Their value depends not only on scientific breakthroughs but also on whether farmers can use them conveniently and effectively," he added.
The university also plans to expand its cooperation with major agricultural production regions, including Heilongjiang, Jilin and Anhui provinces, through research stations, local institutes and science and technology backyards.
These platforms can help researchers understand local production conditions, adapt technologies to regional needs and provide more direct support to rural communities, he said.
Du Jinyi, from the national rural vitalization publicity and education center, said that a workforce capable of operating advanced equipment, interpreting data and managing increasingly sophisticated production systems is in urgent need.
Du said rural talent development programs would need to expand from conventional planting and breeding techniques to cover smart agriculture, digital villages, green and low-carbon production, and practical AI applications.
Training methods should also move beyond classroom instruction toward field-based and hands-on training, while the target group should expand from traditional farmers to include agricultural managers and digitally skilled rural professionals.
The new approach reflects a trend that agricultural modernization can no longer rely primarily on expanding traditional production inputs. Instead, it will increasingly depend on the integration of advanced technologies, equipment, industrial capabilities and skilled personnel.
This shift is expected to accelerate during the 15th Five-Year Plan (2026-30), as China seeks to build a stronger agricultural sector through scientific and technological innovation.
The 15th Five-Year Plan for accelerating agricultural and rural modernization, which was released in June, places particular emphasis on developing emerging and future industries in agriculture, marking the first time such a strategic focus has been explicitly included in the national planning document.
The plan identifies agricultural and rural modernization as fundamental to the overall progress and quality of Chinese modernization, describing it as a cornerstone of the country's modernization drive. It calls for scaling up a range of emerging technologies and industries, including intelligent breeding, new-energy agricultural machinery, the low-altitude economy in agriculture, agricultural bio-manufacturing and novel foods.
Targeted initiatives will also be introduced to upgrade agricultural machinery and equipment as well as promote the integration of AI with agriculture.
An official from the Ministry of Agriculture and Rural Affairs said the initiatives are intended to foster new quality productive forces in agriculture, while strengthening policy support for technological innovation.
Measures will include implementing tax incentives for scientific and technological innovation, as well as policies supporting the first application of newly developed agricultural equipment and products, with local governments encouraged to explore similar support for first-version agricultural technologies and software.
Du noted that the plan's emphasis on agricultural equipment and products, alongside tax incentives for innovation, is significant. "Such policies could help reduce the risks faced by companies developing new technologies and encourage agricultural producers to adopt innovative products at an earlier stage," he said.
Qiu Tongwei, a professor at Nanjing Agricultural University, said emerging agricultural industries are new growth drivers taking shape, while future agricultural industries represent strategic high ground that China needs to secure for tomorrow. "Both of them form the technological foundation of the country's strategy to become an agricultural powerhouse," Qiu said.
He said accelerating the development of emerging sectors such as smart agriculture and the resource utilization of agricultural waste, while cultivating future industries, including gene-editing breeding, would help unlock new momentum for high-quality agricultural development.
Contact the writers at zhaoyimeng@chinadaily.com.cn
