A surgeon places a sheet thinner than a credit card onto the brain and walks away. No threads are pushed into tissue. No robot pierces the cortex. Weeks later, the same sheet is still reading the brain's electrical weather — and then it comes out.
That is the design philosophy of Precision Neuroscience, a New York company founded in 2021 by Benjamin Rapoport, a neurosurgeon and former member of Neuralink's original eight-person founding team, together with business builder Michael Mager. Their wager is that the future of the brain-computer interface (脑机接口) does not require cutting into neurons at all.
A surface, not a spike
Most high-profile implants — Neuralink's threads, Blackrock's Utah Array, Paradromics' micro-wires — sit inside the cortex, among individual neurons. Precision took the opposite road. Its Layer 7 Cortical Interface is a thin-film polyimide array studded with platinum electrodes that lies flat on the brain's surface, conforming to its folds like a second skin.
The bet is mechanical as much as medical. Penetrating electrodes trigger a foreign-body response: glial scar tissue grows around them, signal quality drifts, and recordings degrade over months. A surface array that never breaks the blood-brain barrier hopes to dodge that slow decay. Precision describes the approach as "high-bandwidth, minimally invasive, and safely removable."
The clearance that changed the conversation
In April 2025, the U.S. Food and Drug Administration granted 510(k) clearance to the Layer 7 Cortical Interface for temporary use — up to 30 days — to record, monitor, and stimulate electrical activity on the brain's surface. The FDA's own clearance record (K242618) lists the device as a 1,024-electrode cortical array intended for short-term surgical and monitoring use.
Precision framed the decision as a first: a full regulatory clearance for a next-generation wireless brain-computer interface (脑机接口). That is a company description, not an FDA quotation, but the underlying fact — that the Layer 7 is now an authorized, marketable device for defined temporary use — is documented.
The practical unlock is time. Until now, Precision's arrays had been used only for the few hours of an awake craniotomy. Thirty days changes the math: researchers can collect weeks of high-density neural data from a single implant, and clinicians can map brain function during the critical window after a tumor or seizure surgery.
What the data so far shows
Precision says it has tested the device in 37 patients through partnerships with leading U.S. hospitals — Mount Sinai Health System, the Perelman School of Medicine at the University of Pennsylvania, West Virginia University's Rockefeller Neuroscience Institute, and Beth Israel Deaconess Medical Center. Those implants have so far been limited to short durations during neurosurgical procedures.
The company's near-term product is not a consumer implant but a surgical tool: intraoperative brain mapping. A neurosurgeon removing a tumor near language or motor areas can watch electrical activity in real time and avoid cutting what should not be cut. That is a use case hospitals already pay for, which matters for a company that has raised multiple venture rounds — including a Series C in 2024 — and needs revenue before any permanent implant exists.
The longer game
Precision's stated mission is restoring independence to severely paralyzed people, with stroke rehabilitation and treatment-resistant depression next on the list. The surface array is, for now, the data-gathering front end of that plan. The company argues that neural decoding algorithms, like all AI systems, improve with volume and diversity of training data — and a cleared, longer-duration device is the fastest legal way to accumulate it.
This is the quiet part of the brain-computer interface (脑机接口) story: the bottleneck is not the electrode, it is the dataset. A surface array that can be implanted and removed without permanent risk may simply produce more usable hours of human neural recording than a permanent implant that regulators and patients approach cautiously.
How it compares
Against penetrating rivals, Precision trades resolution for safety. Surface electrodes capture population-level activity across a broad region rather than the firing of single neurons, so fine motor or speech decoding will be harder. But the device avoids the two problems that have dogged invasive implants: tissue damage on insertion and signal loss over time. For the large population that needs some neural interface without brain surgery, that trade may be the right one.
Honest limitations
The cleared Layer 7 is authorized only for temporary, short-term use — not as a permanent implant. The 30-day clearance covers recording and stimulation during neurosurgical care, not at-home communication. Patient-count and partnership figures are company-disclosed; independent peer-reviewed publications on the commercial array are limited. The "first clearance for a next-generation wireless BCI" claim is Precision's own framing. Long-term decoding performance for paralyzed users remains unproven in published form.
What readers can do now
- If you face brain surgery for a tumor or epilepsy, ask whether your center uses high-density cortical mapping — surface arrays like this are already entering the operating room, not just the lab.
- For investors and clinicians, watch clearance scope: the shift from "hours" to "30 days" is the real regulatory milestone, because it enables the datasets every decoder needs.
- For the China angle, the surface-array strategy is where domestic teams can compete without a surgical-robot moat; thin-film manufacturability, not neuron-level penetration, is the skill to watch in the brain-computer interface (脑机接口) field.