BEIJING -- China on Friday launched its first large artificial intelligence (AI) model dedicated to monitoring and protecting arable land, and it will serve as a new digital tool for the national strategy of storing grain in the land and in technology.

Released by the Institute of Agricultural Resources and Regional Planning under the Chinese Academy of Agricultural Sciences (CAAS) on World Soil Day, the model aims to upgrade China's farmland management from quantity-focused preservation to quality improvement -- a task that traditional manual monitoring can no longer handle efficiently.

Developed by a team led by Tang Huajun, an academician of the Chinese Academy of Engineering, the system integrates a central foundation model with dedicated "vertical models" designed for specific tasks such as field segmentation, crop classification and engineering quality inspection.

Wu Wenbin, director general of the CAAS Institute of Agricultural Resources and Regional Planning, said that the AI model goes beyond passive observation. It can diagnose soil health, predict trends and autonomously generate management plans, offering full-life-cycle management for high-standard farmland.

Nationwide, there are more than 66.7 million hectares of high-standard farmland, where soil fertility has been well-preserved with technological advancements and scientific farm management.

In a pilot program in Kunshan, east China's Jiangsu Province, the model successfully created a closed-loop system by adjusting fertilization dynamically based on weather and crop growth cycles.

Industry experts have said the launch provides a technical driver for the sustainable management of agricultural resources, the improvement of arable land quality and the development of high-quality farmland in China.