China’s National Medical Products Administration has granted commercial approval to Neuracle Technology (Shanghai) Co. for a brain-computer interface device - the first such approval anywhere in the world. While brain implants have been used in clinical trials for decades, no country had previously cleared one for general patient use.
The Device and How It Works
Neuracle’s system consists of a coin-sized wireless implant placed on the surface of the brain’s outer membrane (extradural placement, meaning electrodes sit outside the dura mater). The device detects neural signals associated with the intention to move a hand, translates them via software in real time, and sends commands to an air-powered robotic glove worn by the patient.
The key distinction from competing approaches: the implant doesn’t penetrate brain tissue. This reduces surgical risk while potentially limiting signal resolution compared to devices with electrodes embedded directly in neural tissue.
Surgery takes just over an hour. According to Neuracle, patients can typically operate the system independently at home within a month of the procedure. Wireless power and communication allow for long-term use without additional interventions.
Clinical Trial Results
The device was tested in 36 patients. According to China Daily, all participants showed improvement in grasping function - a 100% improvement rate. Some patients also demonstrated what researchers described as “neural remodeling and recovery of additional nerve functions,” suggesting the device may help stimulate neuroplasticity beyond its primary function.
The target population is adults aged 18-60 with cervical spinal cord injuries causing quadriplegia. Patients must have had paralysis for over a year, maintained stable condition for at least six months, and retain some upper arm movement despite being unable to grip objects.
Where This Fits in the BCI Race
China designated brain-computer interface technology as a “national strategic priority” in recent years, and this approval positions Chinese companies ahead of Western competitors on the path to commercialization.
Neuralink, by comparison, received FDA approval for human trials in May 2023 but operates under an investigational device exemption - meaning it can only implant patients as part of approved studies, not sell devices commercially. As of early 2026, Neuralink had enrolled 21 participants across its trials, though the company has signaled plans for high-volume production later this year.
The technical approaches differ significantly. Neuracle’s extradural design prioritizes safety and stability. Neuralink’s penetrating electrodes aim for higher signal resolution, enabling users to control digital interfaces and physical tools through thought alone - a broader use case than the hand-grasp rehabilitation Neuracle targets.
Other Chinese competitors are pushing forward as well. StairMed Technology claims its ultra-flexible neural electrodes are “a fifth the size of Neuralink’s electrodes and hundreds of times softer,” requiring only a 3-5mm skull opening. The company plans approximately 40 patient implants by end of 2026.
What This Means
This approval matters for several reasons:
For patients: People with spinal cord injuries now have a commercially available option in China for restoring some hand function. That’s a genuine quality-of-life improvement, even if the technology remains limited in scope.
For the industry: Commercial viability changes the economics of BCI development. Neuracle can now generate revenue from device sales rather than relying purely on investment capital and trial funding. This may accelerate development timelines across the sector.
For regulators elsewhere: China’s approval establishes a precedent. The U.S. FDA and European regulators will face questions about their own timelines for commercial BCI approvals, particularly as Chinese companies gain real-world deployment experience.
The Fine Print
A few caveats worth noting:
The 100% improvement rate sounds impressive, but “improvement in grasping function” is a broad metric. We don’t have detailed data on the magnitude of improvement, how it varies across patients, or long-term durability of outcomes.
Brain implants can experience degradation over time - scar tissue buildup can degrade signal quality, and devices may shift position. Long-term data from widespread commercial use will be essential for understanding real-world performance.
The device requires patients to wear a robotic glove, which limits its practical applications compared to systems that might one day restore natural hand movement. This is assistive technology, not neural repair.
Finally, commercial approval in China doesn’t mean the device meets regulatory standards elsewhere. Companies seeking U.S. or European approval will still need to complete separate trials and approval processes.
Still, this is a milestone. After decades of brain-computer interfaces existing only in labs and clinical trials, one has now reached the market. The technology works well enough that a major regulatory body deemed it safe and effective for general patient use. That’s worth acknowledging, even as we await broader evidence from real-world deployment.