Inside the Neurotech Revolution: Implants, Interfaces, and Human Potential

Neural Implant Overview

Neural implants, for the purposes of this article, are medical devices implanted in or near the brain, spinal cord, and other major nerves that are intended to provide therapeutic benefit to the patient. There is a fairly long list of such devices, but cochlear implants were among the first neural implants to be approved by the FDA, receiving approval in 1984. They remain the most common neural implant device, with hundreds of thousands of recipients worldwide.

The field of nerve and brain stimulation has grown tremendously over the last 40 years to treat a variety of conditions. For example:

• Cochlear implants – Stimulate the auditory nerve in those with severe sensorineural hearing loss.

• Deep Brain Stimulation – Electrodes implanted in specific regions of the brain connected to a pulse generator. Used therapeutically for Parkinson’s disease, essential tremor, epilepsy, and obsessive-compulsive disorder; experimentally for improving lower-limb movements in patients with severe spinal cord injuries.

• Responsive Neurostimulation – Electrodes connected to a stimulator placed at seizure foci that detect abnormal brainwaves and deliver targeted stimulation. Used for epilepsy.

• Vagus Nerve Stimulation – Electrodes connected to a stimulator attached to the Vagus nerve in the neck. Used for epilepsy and treatment-resistant depression.

• Retinal Implants – Microelectrodes implanted on or near the retina to stimulate cells, allowing partial restoration of vision in certain types of retinal-mediated blindness. This technology is no longer being implanted due to superior newer therapies and company financial struggles.

• Spinal Cord Stimulators – Electrodes placed near the spinal cord and connected to a pulse generator. Used for chronic pain management and, experimentally, to restore movement after spinal cord injury.

Brain–Computer Interfaces (BCIs)

One of the newest breeds of neural implant is intended to facilitate a brain–computer interface (BCI). This technology interprets brain activity and translates it for the purpose of interacting with external devices such as computers, mobile devices, or neuroprosthetics like robotic arms. BCIs can not only read outgoing brain activity but can also stimulate the brain with incoming feedback, for example, providing the sensation of one’s own hand when using a robotic arm to shake hands. This extraordinary ability to both send and receive data allows those paralyzed by spinal cord injuries or neurological diseases such as amyotrophic lateral sclerosis (ALS) to play video games, create art, or use a robotic device to drink water from a cup without a caregiver’s assistance.

More recently, BCIs have garnered considerable attention for their potential to provide those who have lost the ability to speak due to a stroke or neurological disease the opportunity to communicate using only their thoughts. Unlike eye-tracking or other assistive communication technologies currently in use, BCIs can generate conversational vocabulary at a much faster pace and with greater nuance such as word emphasis and tone.

Challenges of Thought-to-Speech

Thought-controlled speech via a BCI is more challenging than thought-controlled external devices due to the inherent complexity of human speech compared with gross motor activity. For example, most thought-to-speech research has required the participant to attempt speech—to actually send the signals necessary for vocalization. Those are the signals evaluated by the implant and translated into words. Unfortunately, some neurological conditions progress to the point where even the attempt to speak is not possible. Further, even if attempted speech is possible, the act can be taxing, much slower than conversational speech, and even physically uncomfortable.

To address this problem, research has expanded to include studies on imagined speech, with encouraging initial results. As with all advances, however, new hurdles arise. Imagined speech brings the legitimate concern that some thoughts are never intended to be spoken aloud yet could be inadvertently translated. A study published in August 2025, however, demonstrated that software can be configured such that the user must imagine a password phrase before their subsequent imagined dialogue is translated. The phrase used in the study was “Chitty-Chitty-Bang-Bang” and was successfully recognized more than 98% of the time. This demonstrates that potential strategies exist for privacy protection, which has long been a concern for researchers of thought-controlled communication.

History & Current State

This sounds like a beautiful use of technology, so what are these devices in practical terms and where does the technology stand from a research and development perspective?

The theoretical and early practical foundation for the current technology has been studied since the late 1990s, and researchers made great initial strides with external devices capturing and interpreting brain activity. Unfortunately, these devices were bulky and could only receive crude, basic brain signals, not the clarity necessary for something as complex as human speech. Processing power at the time was also insufficient to translate complex neural activity and then facilitate an action via an external device without significant latency. However, the explosion of AI and machine learning over the last 5–10 years, coupled with advances in manufacturing and materials, has brought theory much closer to reality.

Companies Developing BCIs

Recognizing that achieving the goal of a brain–computer interface was increasingly possible, a number of companies now have active clinical trials, each angling for recognition in this historic scientific achievement. Some of these companies are high profile. Neuralink is probably the most well-known BCI company, courtesy of its owner Elon Musk. Not to be left out, Jeff Bezos and Bill Gates are key investors in Synchron. Precision Neuroscience was founded by several of Neuralink’s founding members who left one year after Neuralink launched. Other contributors to the space include Blackrock Neurotech (not associated with the controversial BlackRock investment firm), NeuCyber NeuroTech, BrainGate, and others. Each brings a different approach, complete with differing pros and cons, contributing a myriad of valuable findings and experiences to this rapidly growing body of research.

Hardware Differences

One of the primary differences across the companies is the hardware that actually receives and transmits the brain activity. Most devices require brain surgery to place the implant, but even the device itself varies in invasiveness across companies.  Most implants, including the type used by Neuralink, have microelectrodes that actually extend a few millimeters into the brain tissue (image below).  

This punctures the blood brain barrier (BBB) but as a tradeoff also receives a very clear signal.  A key downside to this approach is the fact that puncturing the brain does cause a certain amount of damage, leading to inflammation and scarring which can degrade the quality of the signal and can even lead to the device becoming nonfunctional.  Puncturing the BBB also exposes the brain to unnatural conditions, whether foreign contamination or found naturally within the body such as fluctuations in sodium or potassium that normally would not cross the barrier.

As the brain activity is captured it is sent to a connected “transmitter” (different companies have different name for this portion of the system).  For Neuralink (image below), the transmitter portion of the device replaces a portion of skull and is externally exposed but said to be flush with the skin of the patient’s scalp.  The patient’s use of the device is wireless. 

Other companies like Blackrock Neurotech appear to have transmitter components that extend well beyond the scalp (image below) and are not necessarily wireless when in use.

Precision Neuroscience has a significantly different approach to design.  Their device does not puncture the brain but rather is a film with microelectrodes that lays directly on top of and conforms to it (image below).  The obvious advantage is in the fact that inflammation and scarring associated with puncturing the brain are not a concern.  The approach also allows for more electrodes to be applied to the brain’s surface which provides more neural data and potentially clearer and more comprehensive interpretation. 

Precision Neuroscience's transmitter device (image below) rests between the skull and the scalp, remaining hidden under the skin.  Insertion of the device does technically require brain surgery, but it is minimally invasive via their patented “microslit technique” and no portion of the skull is removed to make space for the transmitter.

Synchron also has a novel design.  The device that collects the neural activity is called a “Stentrode” and looks very much like a heart stent.  

Similarly, it is delivered to its destination during a minimally invasive procedure to access the body’s blood vessels and threading the device to the brain rather than open brain surgery. The transmitter portion of the device is contained under the skin similar to a cardiac pacemaker.  The reduced invasiveness of this device results in a less precise waveform but nevertheless appears sufficient for users to accurately control mobile devices. 

Target Use Cases

Another difference across these BCI companies is the target functionality or use case. While all report ambitious goals, for FDA approval they focus on specific initial applications:

• Synchron – Thought-controlled external devices and digital ecosystems, including integration with virtual reality headsets and apps to control standard mobile device functionality as well as connected peripherals such as ceiling fans or automated pet feeders.  Partnerships with Nvidia and Apple support this vision.

• Precision Neuroscience – Thought-controlled external devices and speech, with planned expansion into BCI for depression treatment and severe motor deficits (as opposed to full paralysis).

• Neuralink – Musk’s stated long-term goal is to facilitate human–AI integration and augmentation (e.g., enhanced vision). Currently targeting three medically therapeutic goals: thought-controlled external devices, thought-controlled speech, and visual prostheses.

• Blackrock Neurotech – Provides complete BCI ecosystems and support for researchers. Its devices have been used in studies aimed at restoring speech, sight, and hearing as well as neuroprosthetics, and digital interaction such as with computers and mobile devices.

• NeuCyber NeuroTech – State-owned Chinese project, focusing on thought-controlled external devices and neuroprosthetics.

Availability Timeline

Each of the companies mentioned has implants in human study participants and has demonstrated enough safety and success to receive clinical trial expansion. Many of the companies’ websites (see references) include patient registries for trial enrollment. That said, the technology is still not ready for widespread use. Time estimates are rare, but most indicate it will take at least several years to come to market. Encouragingly, nearly every company mentioned has received FDA “Breakthrough” designation, which allows for expedited FDA review.

After FDA approval, insurers—an essential part of any device’s journey to market—will have to determine whether the devices merit coverage and who qualifies. A key consideration will be proof of utility. While thought-controlled neuroprosthetics and external devices perform well, they must prove superiority over other technologies such as eye-tracking or voice control.

The ability to communicate is a fundamental human characteristic that is stolen from those suffering from neurodegenerative diseases like ALS or the aftermath of certain types of strokes. As such, BCI for the thought-to-speech use case presents a compelling argument for coverage, since existing technology is extremely limited in comparison. However, from a reliability standpoint, BCI speech technology still has significant hurdles due to the neurological complexities of speech coupled with the software and processing sophistication required for fast, accurate translation.

To be viable, researchers must not only achieve output speed closer to that of natural human speech (roughly 150 words per minute) but also ensure reasonable vocabulary size and acceptable word error rates. While there are exceptions and considerable variability across study designs, most research has utilized limited, pre-defined vocabularies and report word error rates over 25%. Although speech rates also fall short of 150 words per minute, they already well exceed what is possible with current eye-tracking or similar assistive tools.

Finally, devices, regardless of use case, must be made realistic for patient and caregiver use. Currently, some require frequent calibration, monitoring, or maintenance by research staff which is impractical outside of the research environment.

Conclusion

This is a very exciting time for anyone interested in neurotechnology. Recent advances in computing, materials, and manufacturing have resulted in an explosion of potential therapeutic products, with much more on the horizon. Though hardware and software must mature and FDA approvals are still pending, each company in this space brings unique advantages and limitations. This diversity will eventually ensure a range of options for patients to find the devices best suited to their needs and goals. With active trials expanding and already underway, the number of patients with implants will increase dramatically, producing a wealth of data and accelerating refinement. This is an unprecedented race with potentially life-changing benefits for countless patients.

References

Blackrock Neurotech. Empowered by thought. https://blackrockneurotech.com/

BrainGate. Turning thought into action. https://www.braingate.org/

Capoot, A. (2025, April 17). Brain implant cleared by FDA for Precision Neuroscience, a Musk Neuralink rival. CNBC. https://www.cnbc.com/2025/04/17/brain-implant-cleared-by-fda-for-musk-neuralink-rival-precision.html

Chappell, B. (2024, January 30). What to know about Elon Musk's Neuralink, which put an implant into a human brain. WUSF. https://www.wusf.org/2024-01-30/what-to-know-about-elon-musks-neuralink-which-put-an-implant-into-a-human-brain

ColdFusion. (2022). The race to control our brains is heating up [Video]. YouTube. https://youtu.be/CGSFRCeH3dw?si=dhmgzQ94Vx4Q0pYK

Conroy, G. (2025, August 14). A mind-reading brain implant that comes with password protection. Nature. https://doi.org/10.1038/d41586-025-02589-5

Hoeferlin, G. F., Grabinski, S. E., Druschel, L. N., et al. (2025). Bacteria invade the brain following intracortical microelectrode implantation, inducing gut-brain axis disruption and contributing to reduced microelectrode performance. Nature Communications, 16, 1829. https://doi.org/10.1038/s41467-025-56979-4

Kunz, E. M., Abramovich Krasa, B., Kamdar, F., et al. (2025). Inner speech in motor cortex and implications for speech neuroprostheses. Cell, 188(17), 4658–4673.e17. https://doi.org/10.1016/j.cell.2025.06.015

Lapid, N. (2024, August 15). Text-to-speech brain implant restores ALS patient’s voice. Reuters. https://www.reuters.com/technology/text-speech-brain-implant-restores-als-patients-voice-2024-08-15/

Lapid, N. (2024, December 4). Health rounds: Brain stimulation helps restore walking after paralysis in pilot study. Reuters. https://www.reuters.com/world/health/brain-stimulation-helps-restore-walking-after-paralysis-pilot-study-2024-12-04/

Lewington, L., McMahon, L., & Gerken, T. (2025, March 22). The man with a mind-reading chip in his brain—thanks to Elon Musk. BBC News. https://www.bbc.com/news/technology-68667673

Lewis, D., & Winkler, R. (2024, October 31). How brain-computer interfaces could change lives [Audio podcast]. Wall Street Journal Tech News Briefing. https://www.wsj.com/podcasts/tech-news-briefing/how-brain-computer-interfaces-could-change-lives

Mims, C., & Higgins, T. (2025, May 18). This company has a plan to beat Neuralink at the brain-computer interface game [Audio podcast]. Wall Street Journal Tech News Briefing. https://www.wsj.com/podcasts/tech-news-briefing/this-company-has-a-plan-to-beat-neuralink-at-the-brain-computer-interface-game

Neuralink. Pioneering brain-computer interfaces. https://neuralink.com/

Neuralink. (2022). Neuralink show and tell, fall 2022 [Video]. YouTube. https://www.youtube.com/live/YreDYmXTYi4?si=l2czCnpqIphnjh8B

Naddaf, M. (2025, April 10). Brain implant translates thought to speech in an instant: Improvements in computer-interface technology approach natural conversational speed. Nature, 640, 295–296. https://doi.org/10.1038/d41586-025-01162-0

Naddaf, M. (2025, June 11). World first: Brain implant lets man speak with expression—and sing. Nature. Advance online publication. https://doi.org/10.1038/d41586-025-01818-1

Precision Neuroscience. https://www.precisionneuro.io/

Ramsey, N. F., & Crone, N. (2023, August 23). Brain implants that enable speech pass performance milestones. Nature, 620, 954–955. https://doi.org/10.1038/d41586-023-02546-0

Ramsey, N. F., & Vansteensel, M. J. (2025). The expanding repertoire of brain–computer interfaces. Nature Medicine, 31, 31–32. https://doi.org/10.1038/s41591-024-03440-6

Reuters. (2024, November 25). Musk’s Neuralink to launch feasibility trial with brain implant, robotic arm. Reuters. https://www.reuters.com/technology/musks-neuralink-launch-feasibility-trial-brain-implant-robotic-arm-2024-11-25/

Reuters. (2025, May 1). Neuralink’s speech restoration device gets FDA’s ‘breakthrough’ tag. Reuters. https://www.reuters.com/technology/neuralinks-speech-restoration-device-gets-fda-breakthrough-tag-2025-05-01/

Synchron. The brain-computer interface device powering autonomy. https://synchron.com/

Taylor, M. (2024, April 8). Exclusive: Synchron, a rival to Musk’s Neuralink, readies large-scale brain implant trial. Reuters. https://www.reuters.com/technology/synchron-rival-musks-neuralink-readies-large-scale-brain-implant-trial-2024-04-08/

Wairagkar, M., Card, N. S., Singer-Clark, T., et al. (2025). An instantaneous voice-synthesis neuroprosthesis. Nature, 644, 145–152. https://doi.org/10.1038/s41586-025-09127-3

Willett, F. R., Kunz, E. M., Fan, C., et al. (2023). A high-performance speech neuroprosthesis. Nature, 620, 1031–1036. https://doi.org/10.1038/s41586-023-06377-x

Zhang, Z., & Dai, J. (2024). Fully implantable wireless brain–computer interface for humans: Advancing toward the future. The Innovation, 5(3), 100595. https://doi.org/10.1016/j.xinn.2024.100595

Assessed and Endorsed by the MedReport Medical Review Board