The Brain-Computer Interface Race Is Now a Real Industry

For two decades, brain-computer interfaces lived in the same conceptual neighborhood as cold fusion: a field full of TED Talks, full of promise, and almost completely empty of products. That changed in the eighteen months between mid-2024 and early 2026. Three companies — Neuralink, Synchron, and Precision Neuroscience — now have implants in human heads, FDA paperwork in motion, and revenue models that look less like science fiction and more like medical devices.

This is a sector report on what shifted, who is winning, and what 2026 actually looks like for a technology that was, by any reasonable definition, vaporware as recently as 2023.

The Patient Numbers Are the News

Until 2024, fewer than ten humans had ever worn a permanently implanted BCI. By January 2026 the count had crossed fifty across all programs combined. Neuralink alone reports 21 active PRIME-study participants worldwide as of early 2026, including the first U.K. patient implanted under GB-PRIME at UCLH and Newcastle. Synchron has implanted its endovascular Stentrode in more than fifty patients. Precision Neuroscience has run multiple short-term implantation studies under its 510(k)-cleared Layer 7 device.

These are still small numbers in clinical-trial terms. But they represent something the BCI field has not had before: a steady cadence of human deployment, measured outcomes, and adverse-event data accumulating across multiple companies in parallel. That is what an industry looks like in its earliest phase, not what a science project looks like.

Three Very Different Bets

Each of the three leaders has chosen a different surgical and design philosophy. The differences matter because they almost certainly determine which company wins which market.

Neuralink: maximum bandwidth, maximum surgery. The N1 implant places more than a thousand electrodes on flexible threads inserted into cortical tissue by a custom robot. The bet is that piercing the cortex captures dramatically richer signal than any surface electrode can. The trade-off is real surgery, real risk, and a manufacturing pipeline that has to scale a robot, an implant, and a custom OR setup at the same time. PRIME participants have hit 8 bits-per-second cursor speeds and 40-words-per-minute typing — enough for a paralyzed patient to use a laptop without help. The new VOICE study is targeting 140 wpm, which is conversational speed.

Synchron: minimally invasive, blood-vessel deployment. The Stentrode rides into the brain through the jugular vein, lodges in a vessel near the motor cortex, and reads neural signals from the inside of a blood vessel — no craniotomy, no cortical penetration. The signal quality is lower, but the procedure looks like an angioplasty, which means it can be performed by interventional neurologists at thousands of hospitals rather than at a handful of academic neurosurgery centers. Synchron raised a $200M Series D in 2025 explicitly to fund the key trial that should begin in 2026 and lead to the first U.S. premarket approval of a BCI implant.

Precision Neuroscience: surface-only, high resolution, reversible. Precision’s Layer 7 is a film thinner than a human hair that slips through a small slit in the dura and conforms to the cortical surface. It does not pierce the brain. It received 510(k) clearance in April 2025 for short-term clinical use and is now in clinical research at multiple academic centers. The bet here is that high-density surface electrocorticography, paired with modern decoding algorithms, can match or exceed the performance of penetrating arrays — without the risk profile that makes patients (and FDA reviewers) nervous.

The contrast is sharper than anything that exists in, say, the cardiac-stent market. Each design implies a different patient profile, a different surgical specialty, a different reimbursement story, and a different long-term failure mode.

What These Devices Actually Do for Patients

The first commercial market for all three companies is the same: severe paralysis. Spinal cord injury, ALS, brainstem stroke, locked-in syndrome. For these patients, the BCI is not an enhancement. It is a way to operate a computer, send a text, or control a wheelchair when no other channel is available.

The reported outcomes from PRIME and the Synchron COMMAND trial are modest by consumer-tech standards and miraculous by clinical ones. Patients with full hand paralysis are gaming on Steam, posting on social media, writing emails, and controlling smart-home devices. Synchron’s twelve-month safety data showed no serious adverse events across six early-feasibility participants. Neuralink reports stable signals at eighteen months in its first cohort.

A few signals make this look closer to a real product than a research curiosity:

• Insurance conversations are happening. CMS coding committees have begun review of HCPCS codes for implantable BCI procedures, prompted by Synchron’s key-trial timeline. Coverage decisions are still years away, but the regulatory plumbing is being laid now.
• Patients are leaving studies functional. Earlier BCI trials often ended with patients losing access to the device when the study closed. The current generation has plans for long-term continuation, post-trial software updates, and even cross-company portability of decoded signals — all features that imply a product line, not a one-off experiment.
• The decoding stack is improving faster than the hardware. The same advances driving on-device AI are now running inside BCI software, with [smartphone-class chips running LLM-grade decoders without the cloud](https://issuebyte.com/on-device-ai-smartphones-are-running-llms-without-the-cloud/). A patient who got 4 wpm in 2023 can hit 40 wpm in 2026 with the same implant — the upgrade was algorithmic, not surgical.

The Capital and Talent Picture

A useful proxy for whether a deep-tech sector is real is whether serious capital and serious talent are crossing into it. Both are.

Synchron’s $200M Series D came from Khosla Ventures, ARCH Venture Partners, and a strategic position from a major medical-device manufacturer. Precision Neuroscience has raised more than $150M to date. Neuralink’s 2025 round valued the company at roughly $9 billion. These are not seed rounds. They are growth-stage commitments that imply the lead investors believe key trials will read out, FDA approvals will happen, and reimbursement will eventually follow.

On the talent side, the field is pulling experienced regulatory leaders out of cardiac-rhythm management, neurosurgeons with deep-brain-stimulation backgrounds, and machine-learning researchers from large AI labs. That last group is new. BCI decoding is a real-time pattern-recognition problem at heart, and the same architectural advances that made neuromorphic chips a credible challenger for AI inference workloads are now being applied to neural-signal decoding.

Where It Could Still Go Wrong

The risks are real and worth naming honestly.

Long-term electrode stability. Every implanted electrode triggers a foreign-body response. Glial scarring degrades signal quality over months and years. A device that works beautifully at six months and poorly at five years is not a product — it is a recurring surgical bill. The 18-month data is encouraging; the ten-year data does not exist yet for any of these designs.

Surgical scaling. Neuralink’s design requires a custom robot and a specialized OR setup. Even if the device wins on signal quality, scaling to thousands of patients per year is a manufacturing and surgical-training problem the company has not yet solved at industrial scale. Synchron’s endovascular approach mitigates this but at a real cost in signal richness.

Regulatory timelines. Even the most optimistic 2026 timeline still puts the first PMA-approved BCI implant in 2027 or 2028. Class-III device approval is not a fast lane. A negative key-trial readout, a serious adverse event in a pediatric study, or a reimbursement denial could push the field back two or three years.

Indication creep. The current focus on severe paralysis is the right one — clear medical need, clear FDA pathway, clear reimbursement story. The temptation to chase healthy-user enhancement markets (typing speed for office workers, gaming, “augmented cognition”) is real, and historically, that path has destroyed medical-device companies. Watching whether these three founders stay disciplined is one of the more interesting questions in the sector.

The 2026 Verdict

For most of its history, the BCI field has been a graveyard of brilliant prototypes that never reached a tenth patient. The current cohort has crossed that threshold by an order of magnitude. There are now meaningfully different surgical philosophies competing in parallel, real capital deployed against multi-year clinical timelines, and patient outcomes that are solid enough to inform CMS coding committees.

This does not mean a BCI is in your future. It does mean the technology has graduated from the same conceptual tier as flying cars and entered the same tier as deep-brain stimulation in the late 1990s — a real medical device for a narrow patient population, with credible expansion paths into adjacent indications. That is a meaningful upgrade in seriousness for a field that, as recently as 2022, was largely demos and Twitter threads. The next twelve months will produce the first key-trial readouts, and with them, the answer to whether 2026 is the year BCIs stopped being science fiction and started being a product category.