A Moment of Reckoning for Implanted Brain-Computer Interface Studies

The field of brain-computer interfaces (BCIs) has fascinated scientists, clinicians, engineers, and the public since Jacques Vidal¹ coined the term in 1973. Enthusiasm stems from the promise that BCIs might restore lost function, circumvent disability, and shed light on complex aspects of brain function. In recent years, a small number of high-profile companies and academic laboratories have led a surge in implanted BCI studies, bringing with them new commercial aspirations and media attention. Yet, with this expansion comes an imperative to reflect carefully on what these devices offer, to whom, and under what conditions. There is a pressing need for principled, patient-centered guidelines that distinguish clinical promise from speculative or media-driven enthusiasm.

In late 2024, the United States Government Accountability Office released a report outlining policy options around BCIs.² Although the report addressed questions of data ownership, insurance coverage, and long-term device support, it emphasized hypothetical second-order issues such as cybersecurity and data interoperability, while offering scant guidance on the physical morbidity, ethical enrollment, or scientific merit of current human implantation studies.

In response, we aim here to recenter the conversation. We propose a set of considerations and criteria by which BCI studies should be evaluated—focused on participant welfare, clinical relevance, device viability, and integrity of the scientific contributions. Our guiding principle is that BCI research must be not only technologically ambitious but patient-centered, ethically grounded, and scientifically rigorous.

RESEARCH SUBJECT OR PATIENT? A NECESSARY DISTINCTION

A core issue in current BCI research is the conflation of research subjects with patients. This is more than a semantic concern. If a study is not designed to offer a direct or even plausible therapeutic benefit to the individual, that individual is not a patient—rather, they are a research participant in a science experiment. Yet across the BCI literature, media coverage, and even consent processes, this distinction is frequently obscured.

Many studies—especially those involving vulnerable populations such as individuals with amyotrophic lateral sclerosis or high cervical spinal cord injury—implicitly or explicitly present participants as therapeutic recipients when in fact they are engaging with the BCI only during structured research tasks. This distinction matters, and potential misunderstandings can become ethically problematic, risking the specter of exploitation. Participants may describe the experience as positive or uplifting, but these reports must be understood in context. For many individuals with severe disabilities, structured engagement—however limited—may feel rewarding in contrast to otherwise isolating life circumstances. This does not make the intervention itself beneficial. When researchers highlight these narratives as evidence of participant satisfaction, they risk conflating psychological uplift with functional gain. If there is no direct therapeutic link to the participant’s condition, researchers should consider whether healthier volunteers could or should be used—minimizing ethical complexity. The argument that participants “have less to lose” is flawed, especially for the implantation of penetrating electrodes in eloquent cortex, where very limited residual function may be at stake.

WHAT IS THE PURPOSE OF BCI STUDIES?

For implanted BCI research to be ethical and justified, it should offer the following: (1) direct therapeutic benefit to participants; (2) the prospect of generalizable scientific knowledge relevant to future therapies; and (3) a credible pathway toward scalable, sustainable device translation. If a study provides none of these, the primary outcome of BCI work may be limited to media attention and academic prestige. Demonstrations of drone control, avatar manipulation, or social media posting from a brain implant may be headline-grabbing, but do not correlate with learned principles that should guide device development or clinical translation.

USABILITY AND OPEN ACCESS: MOVING BEYOND THE LABORATORY

We believe that all in-human implanted BCI systems must support some form of independent use in the participant’s home, for their own purposes, and without researchers present. To date, most systems operate only within research contexts under technician supervision, never functioning as a clinically relevant prosthetic.

Instead of flashy demonstrations with custom end-effectors, translational BCI software might use brain signals to generate a surrogate to compliment or mimic keyboard, mouse, and eye tracker functions and to obtain back-end control of commercial assistive applications such as Tobii, Wego, AbleNet, and built-in operating system features. These existing assistive devices are valuable because (1) their utility and set of usable features have been validated by use with hundreds of thousands of patients, (2) the rehabilitation infrastructure is already familiar with these devices, and (3) there is already a viable market sustaining these applications.

Moreover, studies must explicitly disclose usability limitations. From the beginning, participants should know not only the risks of surgery but the likely longevity and functionality of the implant and its supporting hardware and software, as well as the duration of technical support provided by the research team. Most penetrating electrode systems in human subjects have lost signal quality within months to years yet are still implanted with the hope of long-term benefit. For implanted BCI research to have real-world impact, it must move beyond bespoke platforms toward device ecosystems that can be manufactured, maintained, and supported at scale. A reasonable path to translation requires not only robust hardware and software but also clinical workflows and provider training that can be standardized across health systems.³ Without scalable infrastructure and implementation expertise, even the most promising technologies will remain confined to research settings, inaccessible to the broader population they aim to serve.

RISK, MORBIDITY, AND PERSONAL BENEFIT: COMPLETE DESCRIPTION AND FULL DISCLOSURE IS ESSENTIAL

Every implanted device carries risk—of infection, hemorrhage, neurological deficit, and long-term morbidity. These risks must be communicated clearly and pragmatically by a noninterested party. One might consider illustrations of chronically exposed hardware, postoperative incisions, and potential cosmetic morbidity should be included in the consent process. Discussions should include the risks of potential explantation, which can exceed those of implantation because of arachnoidal encapsulation, vascular ingrowth, dural or calvarial overgrowth, and incorporation of wiring into the galea and bone. Some study protocols call for routine explantation after trial completion—a practice rarely seen in other implantable therapies. Removing intracranial hardware requires reoperation, sometimes requiring a full craniotomy where only burr holes were required at the time of implantation.

DURABILITY AND DEVICE SUPPORT: WHO IS RESPONSIBLE?

Implanted systems must be sustainable—not only regarding hardware reliability but also software, support, and clinical oversight. Many BCI trials are backed by startups whose longevity is uncertain. If a company fails, participants may be left with orphaned devices. If industry partners cannot guarantee long-term support, investigators and ethics boards must be transparent about this reality during consent. We propose 3 potential contingency measures that may mitigate this issue: (1) connector interoperability mandates, enabling hardware to remain compatible with existing and future alternatives; (2) escrow funds or written hospital pledges, secured at the time of implantation, to cover device support or removal; and (3) predefined sustainability plans, approved by oversight boards, for startup-backed research implants. Trial approvals should include a clear plan for transition to other systems if devices are orphaned, and discussion of this plan should be a component of the consent process.

SCIENTIFIC JUSTIFICATION: TOWARD GENERALIZABLE KNOWLEDGE

Basic science and early-stage device research are essential to the development of BCIs. However, when such studies involve neurosurgical implantation in individuals with severe disabilities, the ethical threshold for justification must be substantially higher. In the absence of direct benefit to participants, the scientific insights gained must be broadly applicable and meaningfully advance the field toward future therapeutic use. Such insights might include identifying which brain regions generate reliable control signals, elucidating how sensory or behavioral feedback shapes cortical activity,⁴ determining which sensor (eg, electrode) scales match each electrophysiological biomarker, what the longevity of biomarkers are,⁵ defining the particular pathophysiological constraints that limit patient groups,⁶ or differentiating various neural correlates of intention.

A study may be ethically justified on the strength of the knowledge it yields—but only if this knowledge is generalizable and contributes to foundational principles of BCI development. Incremental replications of earlier demonstrations, superficial outputs lacking interpretability, or closed systems that withhold data fail to meaningfully advance the field. To have lasting value, BCI research must be grounded in transparency, data sharing, and rigorous methodological design.

We are either constructing the conceptual pyramid of brain-computer interfacing—laying bricks on which future therapies can be built—or we are distributing rubble. Without a commitment to generalizable knowledge, we risk undermining public trust, disappointing medical stakeholders, and provoking backlash from the very communities these technologies aim to serve.

TOWARD PATIENT-CENTERED BCI: A CALL FOR ACCOUNTABILITY AND CLARITY

Implanted BCI studies occupy a critical space at the frontier of neuroscience, technology, and clinical care. As such, they demand a careful and honest balancing of risk and benefit—both to the individual and to society. Ideally, both are served: the participant gains function, and the community gains insight.

A call to action is warranted: to recenter BCI research on the lived needs of people with disabilities, to pursue generalizable principles rather than media-friendly stunts, and to treat implanted participants with the full respect and transparency that medical ethics demand (Table 1). To ensure that we are not simply advancing careers or headlines at the expense of participants, we must commit to the following principles: (1) transparent communication about risk, benefit, and purpose; (2) open access and home usability as baseline expectations; (3) long-term support and interoperability mandates; and (4) a focus on generalizable, scalable, and useful knowledge (Figure 1). By adherence to these principles, true progress will follow—not just in scientific literature or on the news but in the lives of the people these devices are ultimately intended to help.

TABLE. Summary of Issues Facing Researchers in the Field of Brain-Computer Interface Research and Recommendations for Approaching These Issues.

Area of focus	Recommendation
Participant vs patient	Clearly distinguish participants in research from clinical patients. Avoid using therapeutic language unless a clinical benefit is intended
Informed consent	Include explicit discussion of risks (eg, craniotomy, explantation), expected usability, long-term sustainability, and visual depictions of hardware/scars
Device usability and integration	Ensure devices function outside laboratory settings and interoperate with existing assistive technologies used by the target population
Scientific contribution	Design studies to yield generalizable knowledge relevant to future therapies (eg, signal viability, feedback plasticity, cortical integration)
Risk-benefit justification	Justify implantation with either therapeutic intent or significant expected scientific insight. Avoid implantations that serve only as demonstrations
Device sustainability	Require device manufacturers to establish contingency plans for device support and interoperability, especially if the device is not commercially available
Ethical framing	Avoid over-promising capabilities. Ensure media coverage and recruitment materials do not misrepresent potential benefits or blur research and therapy

FIGURE.

Infographic summarizing concerns, transition paradigms, and guiding pillars relevant to the current state of Brain-Computer Interface research.