Brain-Computer Interfaces: From Lab to Clinic to Market

The first brain chip just went on sale
China won the regulatory race — but the real contest is between what these devices can do and how long they last.

On March 13, China's drug regulator did something no country had done before: it approved a brain-computer interface for commercial sale. Neuracle Medical Technology's NEO implant — a coin-sized wireless chip placed on the brain's surface — can now be prescribed to spinal cord injury patients as a standard medical device, not an experiment. A second company, Borui Kang, received approval the same week. The race between Silicon Valley and Shenzhen just found a new arena, and Beijing drew first blood.

Two weeks later, Neuralink reminded everyone why technology matters more than paperwork. Kenneth Shock, an ALS patient who lost his ability to speak, received the N1 Telepathy implant in January 2026. The system learned to decode his intended speech — first from words he spoke aloud, then from silent mouthing, and finally from pure thought. By March, Shock was speaking again, his AI-reconstructed voice matching the sound of who he was before the disease. Meanwhile, another Neuralink patient was playing World of Warcraft by mind control.

The field's ambitions keep expanding. Neuralink's Blindsight device — designed to bypass damaged optic nerves and deliver visual signals directly to the cortex — received FDA Breakthrough Device designation. The first human trial is pending, with initial vision expected at "Atari graphics" resolution that improves as the brain adapts. A Cell Reports paper in April demonstrated speech decoding in a patient with complete locked-in syndrome — the most severe case attempted to date. And in the non-invasive lane, Chinese startup Gestala raised $21.6M for an ultrasound-based BCI that reads and stimulates the brain through the skull, no surgery required.

The historical parallel that matters is the cochlear implant. First implanted in 1961, it took 23 years to get FDA approval (1984), and another 16 to reach mainstream adoption. BCIs just hit their "1984 moment." If the parallel holds, mass adoption is 15–20 years out — but AI-powered decoding, wireless miniaturization, and venture capital ($1B+ for Neuralink alone) could compress that timeline dramatically.

There's a catch, and it's physical. The Utah Array — the 100-electrode workhorse that underlies most invasive BCIs — degrades over time as the brain's immune system encases it in scar tissue. A 2021 review found few studies had rigorously evaluated long-term reliability. New silicon carbide coatings show promise in animal studies, but nobody has solved chronic biocompatibility in humans. The algorithms got smart enough. The electrodes didn't get durable enough. That gap is now the rate-limiting step between a breakthrough and a product.