China approves the world’s first commercial brain implant, leaving Neuralink behind

For the first time in history, a brain-computer interface has crossed the threshold from experimental device into commercial medical product. The technology that achieved it does not come from Silicon Valley or a Musk-backed laboratory. It was built in Shanghai, approved in Beijing, and it works by sitting gently on the surface of the brain rather than plunging into it — a design choice that turned out to be as strategically decisive as it was medically significant.

The device is called NEO, and it is the size of a coin. Developed by Neuracle Medical Technology, a privately held company established in Shanghai, it rests on the brain’s outer membrane without penetrating tissue. Eight electrodes capture electrical signals produced when a patient imagines moving a paralyzed hand. A computer decodes those signals and transmits them wirelessly to a pneumatic robotic glove, which then executes the movement the patient intended. The result, demonstrated across a cohort of 32 patients with cervical spinal cord injuries, is that people who had lost the ability to grasp, hold, or lift objects were able to do so again — picking up cups, eating independently, performing the ordinary transactions of daily life that paralysis had taken from them.

The human reality this device addresses is not abstract. Cervical spinal cord injury strips people of hand function, often permanently, and with it the autonomy that defines adult life. The standard care pathway offers rehabilitation and adaptation, not restoration. NEO represents a fundamentally different proposition: that the brain’s intention to move, still intact and still generating signals, can be intercepted, translated, and returned to the body as actual physical action through an external system guided by thought alone.

Brain-computer interface research stretches back decades, originating in American university laboratories in the 1970s and receiving massive institutional investment under the Obama administration’s Brain Initiative, which directed more than three billion dollars into neuroscience projects over more than a decade. The United States has long held the leadership position in this field, with companies like Neuralink, Synchron, and Paradromics advancing devices of increasing sophistication through clinical trials. Neuralink, founded by Elon Musk, has been the most publicly prominent — inserting thousands of ultra-thin electrodes directly into brain tissue to capture neural signals with extraordinary precision.

That precision comes at a cost. Penetrating electrodes carry documented risks: infection, scarring around the implant, gradual signal degradation as brain tissue responds to the foreign presence. The Food and Drug Administration initially rejected Neuralink’s application for a clinical trial in 2022 before approving it the following year. The trial is still running. As of early 2026, 21 participants were enrolled. No American device has been approved for commercial use.

NEO’s design sidesteps this risk profile entirely. By placing electrodes on the epidural surface — above the protective membrane that surrounds the brain, not inside it — Neuracle’s device avoids the tissue damage associated with deeper implantation. The trade-off is a narrower signal, capturing activity across a region rather than from individual neurons with the specificity that Neuralink achieves. But for the specific application of restoring hand function in spinal cord injury patients who retain some upper arm movement, the epidural signal is sufficient, the glove delivers real results, and the reduced risk profile made commercial approval achievable.

Neuracle submitted clinical data covering up to 18 months of device use. One patient, documented in a preprint published the prior year, used the system at home for nine months and achieved a perfect score on object transfer tests, with a 27-point improvement on the ARAT scale — a standardized measure of upper limb motor function. It is a small dataset by the standards of large pharmaceutical trials, and independent experts have been clear that the cohort of 32 patients is not sufficient to establish long-term safety and efficacy conclusions with confidence. The approval, however, reflects a regulatory judgment that the benefit profile for a population with no other restorative options justifies moving forward.

China’s National Medical Products Administration issued the commercial authorization in March 2026. The timing was not incidental. NEO’s approval arrives embedded in a broader national strategic context: China’s five-year plan for 2026 to 2030 explicitly identifies brain-computer interfaces as a “future industry” — a designation that unlocks coordinated state funding, pricing frameworks, and institutional priority across the research and healthcare system. China’s National Healthcare Security Administration had already released guidelines establishing standardized fee codes for BCI implantation procedures, creating the reimbursement infrastructure before a single commercial device existed. The policy architecture was ready. The device filled it.

The geopolitical dimension of this development cannot be separated from its medical significance. Xi Jinping has described the technology sector as the primary battlefield of global competition, and the brain-computer interface industry sits at the intersection of neuroscience, artificial intelligence, advanced manufacturing, and national security — making it precisely the kind of domain where China’s model of coordinated state-commercial strategy produces outcomes that market-driven systems moving through independent regulatory processes cannot easily match for speed.

Independent researchers have assessed China’s BCI capabilities as broadly comparable in sophistication to those of the United States and the United Kingdom. The gap is not that one side has capabilities the other lacks — both have serious science, serious capital, and serious ambition. The gap is in regulatory architecture and strategic alignment. The FDA’s process, built around rigorous independent safety review, is slower by design. It is designed to protect patients from premature commercialization of inadequately tested devices. Whether NEO’s approval reflects a genuinely appropriate risk assessment or a regulatory process that moved faster than the science fully supports is a question that the next several years of real-world patient data will answer.

What is not in question is the competitive reality that China’s approval has created. Neuracle Medical Technology is now the only company in the world with a commercially approved brain-computer interface. Its device is reaching patients outside of trial conditions. The knowledge generated from those patient interactions — the data on long-term use, on device performance in real home environments, on the complications and adaptations that emerge outside controlled clinical settings — will compound. More patients means more data. More data means faster iteration. The commercialization lead, even if built on a less technically ambitious device, creates structural advantages that are difficult for trial-stage competitors to overcome quickly.

For the millions of people living with paralysis caused by spinal cord injury, the immediate meaning of this development is more elemental than any geopolitical calculation. A technology exists that can restore some of what was lost. It has been approved for use. The question of who builds the next generation of that technology, and who controls the clinical infrastructure around it, belongs to a longer race that has now, unmistakably, begun in earnest.