Chinese Researchers Achieve Advanced Physical System Control via Implanted Brain-Machine Interfaces

Researchers in China have announced a significant advancement in brain-machine interface (BMI) technology, demonstrating the capacity for thought-controlled manipulation of complex physical systems. The achievement, publicized in December 2025 by the Center for Excellence in Brain Science and Intelligence Technology (CEBSIT) of the Chinese Academy of Sciences (CAS), involved two patients in their 30s successfully operating external devices using only neural signals from implanted systems.

This development moves the field beyond earlier non-invasive or simple cursor-control interfaces, signaling a major step toward functional restoration for individuals with high-level paralysis. The trials, conducted with participants at Shanghai Huashan Hospital, showcased control over a power wheelchair, a robotic dog retrieving deliveries, and a robotic arm grasping a cup for drinking water. Key performance metrics reported by the research team include achieving a low latency of under 100 milliseconds between the initiation of a thought and the corresponding machine response. Furthermore, one participant demonstrated the ability to control these devices just five days following the surgical implantation procedure.

The collaborative effort involved personnel from CAS, CEBSIT, and specialists from Shanghai Huashan Hospital, including scientists such as Zhao Zhengtuo and Professor Li Meng. This technological progression places China as the second nation, following the United States, to advance invasive BCI technology into this advanced clinical trial stage. The researchers project that this advancement will pave the way for restoring complex motor and speech functions within the next three years.

The physical hardware underpinning this success is reportedly a small, invasive system designed for practicality, contrasting with earlier bulky instruments. While earlier invasive trials in China, such as one in March 2025 involving a tetraplegic patient controlling games, focused on digital interaction, these new cases integrate the physical environment. The system's low latency is critical for real-world interaction, and the robotic arm utilized is noted as a cost-effective model compared to systems developed by entities such as Neuralink. The research aims to promote large-scale applications for individuals with disabilities, potentially offering a breakthrough for millions suffering from complete spinal cord injuries, bilateral upper limb amputations, and amyotrophic lateral sclerosis (ALS).

Contextually, the development of the physical interface has been supported by parallel innovations within the Chinese scientific community. Researchers from the Institute of Automation under CAS have independently developed specialized robotics capable of implanting ultra-fine, flexible microelectrodes into animal brains with micron-level precision. Such flexible electrodes are designed to move naturally with brain tissue, minimizing damage and promoting the long-term stability essential for practical clinical use. One patient is already leveraging brain control to work as an intern product sorter, verifying the accuracy of artificial intelligence systems used in vending machines.