JD.com to deploy 3 million robots to fully automate logistics

Chinese e-commerce titan JD.com is ramping up its robotics ambitions with a bold plan to automate its extensive logistics network, while simultaneously retraining its extensive courier workforce for technical roles. During a Beijing event on Wednesday, JD Logistics unveiled its industrial Wolf Robot series, marking a significant step towards a fully automated supply chain.

Chairman Cao Peng announced the company's goal to acquire 3 million robots, 1 million unmanned vehicles, and 100,000 delivery drones over the next five years. JD Logistics' Wolf Robot system is designed to handle repetitive, labor-intensive tasks with minimal human involvement, picking, sorting, transporting, and delivering goods.

Specialized units can operate in extreme conditions, including temperatures as low as minus 20 degrees Celsius. The company also unveiled delivery drones and automated pharmacy dispatchers as part of its automated solution lineup. The push towards automation is driven by JD and its ecosystem companies employing approximately 700,000 delivery and logistics personnel.

In response to potential job displacement concerns, JD's founder Liu Qiangdong introduced the Nirvana Plan in earlier this year. This initiative aims to retrain couriers and warehouse staff for technical roles such as robot servicing and maintenance, ensuring frontline workers remain valuable as automation progresses. Beyond hardware, JD believes its edge in embodied artificial intelligence (AI) lies in data collection.

To gather sufficient data for training general-purpose embodied AI models, JD plans to leverage its extensive network of warehouses, retail stores, and pharmacies, aiming to collect more than 10 million hours of real-world operational data. The company also intends to establish over 80 RoboBase hubs across China, focused on robotics research, pilot production, data annotation, equipment upgrades, and maintenance.

To support the computational demands of this ecosystem, JD signed a deal with China-based chip designer Moore Threads to construct a computing cluster powered by 100,000 domestic chips. This cluster will be used to train and run AI models, as well as embodied AI applications, and will help JD integrate domestic chips into an AI development pipeline, encompassing data collection, model training, simulation, and deployment.