Digital News

　　In the greenhouse, a “steel-structured” robot shuttles effortlessly among the flowers, accurately identifying each bloom and extending its robotic arms to gently complete hybrid pollination, handling the full breeding process previously done by humans.

　　A scene once limited to a science fiction movie has now become a reality. On the evening of August 11, the world’s first full-process intelligent breeding robot, “GEAIR,” was featured in the international journal Cell, highlighting China’s pioneering achievement in the deep integration of biotechnology and AI in the agricultural intelligent breeding sector. This milestone marks a shift in agriculture from “experience-driven agriculture” to “precision industry.”

　　Overcoming Bottlenecks in Hybrid Breeding and Seed Production

　　Compared with the process involved in producing well-known hybrid rice varieties, GEAIR can apply new-generation breeding technologies, such as de novo domestication and speed breeding, to achieve rapid customized breeding of superior varieties. For instance, in soybean breeding, it has for the first time enabled the rapid development of stigma-exserted male sterile lines, which is expected to break through a bottleneck in soybean hybrid breeding and significantly increase per-unit yield.

　　According to Xu Cao, a researcher at the Institute of Genetics and Developmental Biology (IGDB) of the Chinese Academy of Sciences (CAS) and corresponding author of the study, this study has initiated an intelligent breeding model of “integrated biotechnology, AI, and robot labor.” It marks China’s successful pioneering efforts in the construction of a closed-loop technology system for intelligent robotized hybrid breeding. It also shows the application prospects of “AI for science” in the sector of biological breeding.

　　Reviewers for the journal Cell highly praised the study, describing it as an exciting and innovative breakthrough with broad application prospects.

　　“Customized” Intelligent Breeding

　　Hybrid breeding, which seeks to ensure the “superior birth and raising” of crops, is an important way to enhance both crop yield and quality. By combining superior genes from different parents, such as disease-resistant genes from the male parent and high-yield genes from the female parent, the offspring can inherit the best traits from each, exhibiting “hybrid vigor” that surpasses both parents.

　　“Hybrid breeding and seed production usually involve first removing the male parts of the female parent and then carefully applying pollen from the male parent onto the stigmas of the female parent. The entire process depends on skilled workers working within a narrow window before the flowers bloom,” Xu Cao explained. In hybrid breeding, this procedure must be repeated with different parent combinations, while in hybrid seed production, it is performed on a large scale with fixed parent lines, making the process time-consuming and labor-intensive.

　　Taking tomatoes as an example, hybrid varieties now make up over 90% of commercial varieties. Yet, because of their closed flowers with inward-facing stigmas, tomato hybrid breeding and seed production worldwide still depend entirely on manual pollination. Labor costs for this work exceed 25% of the total tomato breeding expenses, with manual emasculation alone accounting for 40% of the costs of hybrid pollination.

　　“More importantly, some crops with closed flowers, like soybeans, are still unable to harness hybrid vigor due to the high costs of hybrid seed production,” Xu Cao explained. By using gene editing to reshape the flower structure, making the stamens naturally dehisce and sterile while exposing the stigmas, researchers have essentially created a “customized interface” for robots. Coupled with AI-driven visual recognition and advanced positioning technologies, this enables robots to navigate efficiently among the plants and carry out pollination, greatly enhancing breeding efficiency.

　　Unlocking Infinite Possibilities for the Future of Agriculture

　　From Academician Yuan Longping’s pioneering “three-line” hybrid rice breeding system to the recent development of the world’s first full-process intelligent breeding robot, “GEAIR,” generations of Chinese scientists have dedicated themselves to the frontlines of agricultural production. Through technological innovation, they have pushed beyond the limits of farmland and natural conditions, achieving groundbreaking results in the field.

　　“When biotechnology joins forces with artificial intelligence, it can unlock limitless possibilities,” said Yang Minghao, co-first author of the study and an associate researcher at the Institute of Automation, Chinese Academy of Sciences (CAS). He noted that this intelligent breeding model is expected to be applied across a broader range of scenarios and spaces in the future.

　　“For example, robots are better suited for operating in enclosed environments, offering breeding solutions for building greenhouses on the Moon or even Mars, and laying a solid foundation for space agriculture. Likewise, for wild crop species at risk of extinction, rapid domestication techniques paired with this new intelligent breeding model could provide a novel approach for conserving genetic resources,” said Yang Minghao.

　　“When robots can ‘read’ every single flower, breeding will transition from ‘experience-driven agriculture’ to ‘precision industry,’” Xu Cao said. He emphasized that only by prioritizing innovation and relying on science for methods and solutions can China secure its crop production more firmly and reliably.