In 2016, Oxford neurosurgery researchers coined the term “brainjacking” to describe an attacker seizing unauthorized control of an implanted neural device. Their review focused on deep brain stimulation implants, a device already used to treat conditions like Parkinson’s disease by sending electrical pulses into the brain. They catalogued attacks ranging from draining the device’s battery and shutting off therapy to deliberately altering a patient’s mood, impulse control, or pain perception. These effects have no real equivalent in an ordinary data breach, since here the attack surface is a person’s own nervous system.
In December 2025, Elon Musk announced that Neuralink would move to “high volume production” of its brain computer interface in 2026, alongside a shift to an almost fully automated surgical procedure. A brain computer interface, or BCI, is a chip implanted in the brain that translates neural activity into digital commands. The company has grown from a single implant recipient in early 2024 to 21 patients by February 2026, and Carolina Aguilar, CEO of competitor INBRAIN Neuroelectronics, has estimated that “high volume” realistically means hundreds to low thousands of implants a year in the near term, rising toward the tens of thousands in the longer term. This is the leading edge of an industry moving toward population scale deployment, carrying a cybersecurity problem that researchers have flagged for a decade without resolving it. For NATO, these devices are entering its own defence health and veteran care systems, turning an unresolved security gap into a hybrid threat problem for the Alliance.
The risk extends beyond the implant itself to the wireless connection the device uses to communicate, typically Bluetooth. Research has catalogued spoofing and interception vulnerabilities in Bluetooth enabled BCIs, including a known flaw called BLESA that lets an attacker impersonate a trusted device and potentially intercept neural data or interfere with how the implant functions, echoing the mood, motor control, and therapy-disabling effects cited above. In plain terms, the same wireless weakness that can affect a phone or headset can, in a BCI, mean tampering with a signal running to and from someone’s brain.
Regulatory oversight has not kept pace with either strand of this risk. A 2025 review in Frontiers in Human Neuroscience concluded that ethical review of implantable BCI research remains well behind the pace of the technology’s clinical advance, citing unresolved questions of privacy and patient autonomy distinct from the aforementioned brainjacking and spoofing risks. The review flags two recurring gaps: privacy over raw neural data, which can reveal far more about a person’s mental state than a conventional health record, and patient autonomy, since a device that can alter mood or motor control raises consent questions that ordinary surgical consent forms were never built to cover.
Nor do the instruments meant to govern device security close the gap. The United States’ Food and Drug Administration (FDA) premarket cybersecurity guidance, the international standard IEC 81001-5-1 on health software and IT network security, the European Union’s Cyber Resilience Act, and harmonized international cybersecurity principles treat medical device security as a data protection and functional safety matter. None of them address the brainjacking scenario, where an attacker is manipulating cognition, emotion, or motor control through the device itself, largely because these frameworks predate implantable BCIs and were built for conventional software and data breach risks. That gap has a direct consequence for informed consent: a patient, or a servicemember, cannot meaningfully consent to a risk that no disclosure framework currently requires anyone to name.
It is this scale of civilian exposure that turns a medical device question into a hybrid threat question. NATO’s 2021 “Foster and Protect” strategy on emerging and disruptive technologies already treats dual use technologies, meaning ones with both commercial and defence relevance, as central to Allied resilience, and explicitly recognizes such technologies can be exploited by state and non-state actors below the threshold of armed conflict. A neural interface with a documented wireless attack surface, deployed across tens of thousands of civilians in Allied states, fits this category: it is commercially driven, defence relevant through veteran and battlefield medicine, and exploitable in ways that could unsettle public confidence or target specific individuals – for instance, extracting neural data for blackmail or degrading a target’s mood – without ever triggering a traditional security response.
NATO does not need to wait for a civilian scale incident to justify acting, as it already possesses a narrower and more concrete point of entry via its servicemembers and veterans. Neural interfaces are positioned for rehabilitative use in combat injury and neurological injury care, so NATO member states will be adopting this technology into defence health and veteran care systems on a predictable timeline, under a duty of care NATO already holds toward this population. This provides the Alliance the standing and the institutional machinery, military medical ethics review, defence health procurement, and standardization bodies to act sooner rather than reacting once the same vulnerability has been exploited at civilian scale.
Concretely, NATO should direct its Science and Technology Organization, working with the NATO Standardization Office, to establish an ethics integrated Neural Device Security Standard governing any BCI entering Allied defence health or veteran care systems. This should be framed as a resilience and hybrid threat measure, not a servicemember protection measure, and it should rest on four parts, each enforced through NATO procurement practices, as disclosure and standards ultimately depend on private sector cooperation.
First, technical certification: independent verification, by accredited NATO or national testing laboratories, against a cybersecurity baseline covering wireless transmission security and resistance to spoofed signal attacks, required before any device is adopted into an Allied program.
Second, consent specific disclosure: a standardized cyber risk disclosure, reviewed by military medical ethics boards alongside existing informed consent processes, telling servicemembers and veterans what is, and is not, protected against, closing the gap left by existing regulatory frameworks, so implantation consent never doubles as consent to an unstated cyber risk.
Third, a feedback loop into global standards: NATO should feed its technical and ethical findings into the next revision cycle of the international harmonized cybersecurity principles, issued through the International Medical Device Regulators Forum’s Cybersecurity Working Group, where Allied regulators including the FDA and European Commission hold seats and meet regularly. This builds Allied consensus into the global standard before BCIs reach mass civilian adoption.
Fourth, these same findings should also reach civilian resilience structures, routed through NATO’s Resilience Committee and national civil preparedness mechanisms, so lessons drawn from the servicemember and veteran population through early warnings, incident investigation, and data on implant behaviour under attack inform civilian regulators before a population-scale incident.
Taken together, this sequence uses NATO’s existing servicemember and veteran population as a proving ground for a standard that Allied governments, standards bodies, and manufacturers will need regardless, and it pre-empts a problem that is accelerating faster than any regulator is tracking it. Neuralink’s 2026 timeline is the clearest signal that the window for building this standard proactively, rather than reactively, is closing. What NATO chooses to do with its own servicemembers in the coming years will likely determine whether cognitive liberty is protected by design or defended only after the fact.
Image Citation: “A person’s head with a circuit board in front of it” (2023), Steve A Johnson via Unsplash.
Disclaimer: Any views or opinions expressed in articles are solely those of the authors and do not necessarily represent the views of the NATO Association of Canada.
