The Race for the Brain: Neuralink, China, and the Question of Where Europe Stands

In March 2026, China became the first country in the world to approve a commercial brain implant — a milestone Neuralink has yet to achieve. At the same time, the market for brain interfaces raised hundreds of millions in the first months of 2026 alone. A tour of the players in the US and China, and the uncomfortable question of what Europe is actually doing.

---

In January 2024, a paralyzed man named Noland Arbaugh became the first human with a Neuralink implant. Within weeks, he was playing online chess using only his thoughts; within months, he was posting on social media about life with a chip in his motor cortex. That moment did for brain interfaces what AlphaGo did for AI in 2016: it turned a niche field into something the general public realized was actually happening.

Two years later, in 2026, that field has matured beyond recognition. There are dozens of people with implants spread across multiple companies, genuine regulatory pathways have been established with watchdogs, and there are three or four serious technological paradigms. The brain-computer interface (BCI) is no longer science fiction — it has become a capital-intensive commercialization race with a sharp geopolitical dimension.

Time for an overview: who are the players, where do the US and China stand, and what is the role of Europe?

What a BCI actually is — and the core trade-off

A brain interface picks up signals from the brain and translates them into an action: moving a cursor, controlling a robotic arm, generating speech. The entire field revolves around one central trade-off: the deeper you go into the brain, the better the signal — but the greater the surgical risk.

That trade-off defines the entire map:

• Penetrating (intracortical). Electrodes go deep into the brain tissue. Highest signal quality, highest risk. This is Neuralink's route.
• On the surface (epidural/subdural). Sensors rest on or just beneath the dura mater, without touching tissue. Lower risk, slightly lower resolution. The route of the Chinese NEO and of Precision Neuroscience.
• Endovascular (via the blood vessels). A stent-like mesh is pushed through a blood vessel to just near the motor cortex. No skull opened, no incision in the brain. The route of Synchron.
• Non-invasive. EEG headsets, or newer techniques using focused ultrasound. No surgery, but a weaker and more noise-sensitive signal.

None of these routes is "the right one". Each player chooses a different point on that line — and that explains why the competition is not a matter of who wins, but of who solves which problem.

The United States: spectacle and diversity

Neuralink is the most recognized name in the field. The company is running two main trials: PRIME (cursor and robotic arm control for paralyzed individuals) and VOICE (speech decoding for people with severe speech impairments). The N1 implant is about the size of a coin (23 mm) and contains up to 3,072 electrodes. By early 2026, Neuralink had implanted around 21 participants across the US, the UK, Canada, and the UAE; by mid-year, reports circulated about a 26th patient — a sign that the pace is accelerating. So far, the company reports zero serious device-related incidents, raised a $650 million Series E round in June, and is now valued at roughly 8 to 9 billion dollars. Musk announced "high-volume production" for 2026 along with an almost fully automated implantation procedure.

But American strength lies precisely in the diversity surrounding it:

• Synchron chooses the endovascular route with its Stentrode — no skull surgery. The company raised a $200 million Series D round in late 2025 (with investors such as Bezos Expeditions and the Qatar Investment Authority) and is aiming for a pivotal trial in 2026. If successful, Synchron could be the first to submit a premarket approval (PMA) application for a permanently implanted communication BCI — a milestone the field has been chasing for decades.
• Precision Neuroscience balances channel count and safety with a wafer-thin surface implant (Layer 7) and received 510(k) clearance for temporary implantation for up to 30 days.
• Paradromics focuses on high data bandwidth for speech restoration and received FDA approval for its Connect-One trial in late 2025.
• Blackrock Neurotech is the silent workhorse: its Utah Array has been in dozens of academic patients for over ten years and thus provides the most cumulative hours of human neural data of anyone.
• Merge Labs (co-funded by OpenAI, valued around $850 million) and Kernel focus on the non-invasive and data infrastructure side.

The American model is, in a word, innovation-driven: flexible, capital-rich, with the FDA accelerating the route to market via "Breakthrough Device" designations. According to sector analysts, in the first months of 2026 alone, more than $650 million in new capital flowed into the US and Chinese BCI sectors; at this rate, 2026 will end up at around $2 billion.

China: first to market — with a different strategy

Here lies the real news of 2026. On March 13, the Chinese regulator NMPA became the first in the world to approve an invasive BCI for commercial use: NEO, developed by Shanghai-based Neuracle Technology in collaboration with Tsinghua University. This is no longer a clinical trial — it is an approved product, on its way to being included in the regular reimbursement system. That last point is exactly what distinguishes a commercial product from a trial.

NEO is a coin-sized implant with eight sensors that rest on the dura mater, without touching brain tissue. The procedure takes about 90 minutes. In trials, all participants showed improvement in their grasping ability; some even showed signs of neural remodeling. After training, one participant was able to independently pick up a cup and drink using mind control.

The fact that China reached the market earlier than Neuralink is not because it is more advanced technologically — it is a deliberate strategic choice. NEO opted for the less risky surface route: lower risk profile, fewer regulatory hurdles, simpler surgery. The price paid is signal resolution. In short, China opted for safe and to market first, whereas Neuralink opts for maximum bandwidth.

And there is more to come:

• NeuCyber NeuroTech (state-backed, affiliated with the Chinese Institute for Brain Research) is developing the Beinao series and aims for 50 implantations by the end of 2026. The company claims to be "only three years" behind Neuralink — and believes it can catch up.
• NeuroXess is running a year-long trial with a "triple-full" system: fully implanted, fully wireless, fully battery-integrated.

State support is crucial. China has designated BCI as a strategic emerging industry, with policy support dating back to the 13th Five-Year Plan. The 2026 government work report and the 15th Five-Year Plan explicitly prioritize BCI. The Chinese BCI market was already worth around $446 million in 2024. The Chinese model is state-led and safety-first — with an industrial chain that covers the entire field, from components to applications.

A striking, underreported detail: despite the geopolitical climate, the American and Chinese BCI industries are partially collaborating. The American company Axoft is testing its own BCI in China, in partnership with a Chinese company and a Shanghai hospital.

And Europe? Strong science, weak scale — but a different weapon

This is where the story gets most interesting for Dutch readers. Europe has no Neuralink and no NEO. What it does have:

• CorTec (Germany, a spin-off of the University of Freiburg) was the first to put an implantable German BCI into a human with its Brain Interchange system, received FDA clearance, and in April 2026 earned a Breakthrough Device designation for stroke rehabilitation — backed by signal stability data spanning 500 days.
• INBRAIN Neuroelectronics (Spain) relies on graphene — a flexible, high-resolution material that can replace metals like platinum for durable implants. It reported the first results of a first-in-human study during tumor surgery.
• Onward Medical (Dutch-Swiss) is building a "digital bridge" for paralysis: it bridges a severed spinal cord with targeted stimulation and received 510(k) clearance for home use of its ARC-EX system in late 2025.
• Newronika (Italy) has a CE-marked closed-loop deep brain stimulation system, AlphaDBS, for conditions like Parkinson's.
• Additionally, there is a broad foundational layer: Neurosoft Bioelectronics, MindMaze, CoMind, the Wyss Center (which is working with UMC Utrecht on a fully implantable BCI for locked-in patients), and public research groups. According to databases, Europe is home to around 77 BCI startups, of which roughly 40 are funded.

The problem is scale and capital. Whereas Neuralink alone is worth around $9 billion in valuation, European players raise rounds of a few tens of millions — INBRAIN's most recent injection was even a €4 million grant rather than a venture round. Europe has the science, but lacks the venture capital engine and the unified market to convert it into dominant companies.

The European counterweight: rules and rights

But Europe is playing a different game. Where the US lets innovation lead and China the state, the EU opts for an empowerment model: strictly limiting risks and putting rights first. That is not an afterthought, but potentially Europe's greatest influence on the world stage.

The building blocks are already there:

• The EU AI Act classifies brain interfaces as high-risk AI and explicitly bans the use of AI and neurotech to "rewrite someone's mind" — behavior-altering systems that bypass rational control are considered an unacceptable risk.
• The Medical Device Regulation (MDR) imposes strict requirements on implants via the CE mark.
• The discussion on neurorights — mental privacy, cognitive freedom, ownership of neural data — is further along in Europe than anywhere else. Not everyone agrees that new rights are needed: some experts argue that existing fundamental rights to freedom of thought, privacy, and integrity are sufficient if properly interpreted. But the direction is clear.
• At the global level, UNESCO adopted a framework in November 2025 looking to treat neural data as a special category — "the final frontier of privacy". Chile has already embedded neurorights into its constitution; American states like California and Colorado have included neural data in their privacy legislation.

The danger is fragmentation. BCI regulation is currently divided across medical device legislation, AI rules, data protection, and emerging neurorights — without a single cohesive standard. That creates "regulatory arbitrage" and unequal protection. Think tanks such as the Centre for Future Generations therefore advocate for a separate EU neurotech strategy, so that Europe does not have to choose between strong innovation and robust safeguards, but can combine both — before market dynamics and geopolitical competition make the choices for Europe.

The uncomfortable core: mental privacy

Behind all the company names and funding rounds lies one question that defines the entire field. Every legal system in the world protects the privacy of what you say, write, and send. No legal system adequately protects the privacy of thinking itself — simply because, until now, thoughts have been inaccessible.

Neural data is uniquely personal: unlike a fingerprint, it can reveal thoughts, emotions, or cognitive states that someone never intended to share. Research shows that a disturbing portion of consumer neurotech terms and conditions permit the resale of anonymized brain patterns to third parties. As soon as this technology moves from the operating room to wearables, gaming headsets, and the workplace — and that is already happening — mental privacy will no longer be an academic issue, but a concrete corporate risk and human right.

What this means

The 2026 BCI race is essentially a reflection of three philosophies. The US gambles on innovation and capital, delivering the most spectacle and the widest range of approaches. China chose the pragmatic route, thereby capturing the global scoop of the first commercial implant. Europe has the science and the ethical head start, but lacks the scale — and will exert its influence through rules and rights rather than through dominant companies.

For those working with technology and data in Europe, the lesson is not to try to win the hardware race, which is probably unwinnable here. The lesson is that neural data becomes the next frontier of data protection, that regulation (AI Act, MDR, neurorights) is taking shape right now, and that Europe's credibility lies in building trust where others are mainly building for speed.

We are, as one analyst aptly put it, in the "dial-up phase" of neurotech: clunky and early, but with a clear view of a future where the boundary between biological brain and digital device blurs. The question is no longer whether that future is coming, but who will write the rules of the game.

// About the author

Remy Gieling

Mede-oprichter, AI-expert & bestseller-auteur

Tech-expert (1988) gespecialiseerd in kunstmatige intelligentie en mede-oprichter van ai.nl, The Automation Group, Proxies en eBrain.ai. Oud-hoofdredacteur van diverse zakenmerken en daardoor een geoefend verteller op het podium en in de media. Verzorgt jaarlijks 150+ AI-keynotes in binnen- en buitenland en is gastdocent aan Nyenrode. Co-auteur van zeven boeken, waaronder 'Handboek AI Strategie' en 'AI Agents', en bekend als presentator op radio en RTL Z. Reist langs de labs van OpenAI, Nvidia en Tencent en vertaalt de nieuwste doorbraken naar inzichten die leiders direct kunnen toepassen.

LinkedIn