A sensor smaller than a contact lens, implanted at the surface of the brain and capable of translating the thought of moving a hand into a command for a computer, a robotic arm, or a communication device — that is BrainGate, one of the most closely studied brain-computer interfaces (BCIs) of the past two decades. Emerging from US university laboratories in the late 1990s, BrainGate has opened a direct channel between the brain and the wider world for people with severe paralysis. Behind the technology sits a portfolio of more than 30 patents that has outlasted the company that first built it — passing through several owners over fifteen years before being donated in its entirety to Tufts University in 2020.
From university research to a pioneering neurotechnology spin-off
The research foundations for BrainGate were laid across a series of investigations at Brown University, Columbia University, and Harvard University throughout the 1990s. John Donoghue, the founding chair of Brown University's Department of Neuroscience, led key work on how the motor cortex encodes voluntary movement, establishing that the electrical signals of individual neurons could be recorded, decoded, and used to direct external devices.
To bring these discoveries into clinical use, Donoghue co-founded Cyberkinetics Neurotechnology Systems, Inc. (Cyberkinetics) as a Brown University spin-off based in Foxborough, Massachusetts. A first funding round of US$9 million, led by Oxford Bioscience Partners, provided early capital, and Cyberkinetics subsequently raised more than US$40 million for research and development before going public through a reverse merger with Bionic Technologies Ltd in 2004. In the same year, the US Food and Drug Administration (FDA) granted Cyberkinetics the first of two Investigational Device Exemptions (IDEs) to begin clinical trials on human patients, with sites in Rhode Island, Massachusetts, and Illinois enrolling four participants with tetraplegia.
How the BrainGate system reads the brain
The BrainGate Neural Interface System works in two stages: signal acquisition in the brain and signal decoding outside it.
The core component is a microelectrode array — a small square of silicon containing 100 hair-thin electrodes, smaller than a contact lens — surgically placed on the surface of the motor cortex. There, the electrodes record the electrical activity of individual neurons as the user thinks about moving. The resulting signals are transmitted to an external decoder: software and hardware that converts these neural patterns into commands for an external device, which might be a desktop computer, a powered wheelchair, a robotic arm, or a communication system. No physical movement is required. The BrainGate system reads the intent and acts on it.
In an early trial, a participant paralyzed from the neck down was able to steer a motorized wheelchair, open and read emails, play video games, and control lights and a television — entirely through thought. "Our ultimate goal is to develop the BrainGate System so that it can be linked to many useful devices," said Donoghue. The technology was designed to assist people with tetraplegia, ALS, brainstem stroke, and other conditions causing severe motor impairment. Asked whether he wanted to walk again, one early participant replied simply: "No, I'd just like to be able to scratch my own nose."
Two decades of clinical milestones: from cursor control to speech
The BrainGate research program has grown from a single-institution clinical trial into a multi-institutional consortium spanning Brown University, Massachusetts General Hospital, Stanford University, and the Providence VA Medical Center, among others. It is directed by Leigh Hochberg, a professor of engineering and brain science at Brown and a critical care neurologist at Massachusetts General Hospital.
Progress has been substantial. In 2012, a paper published in Nature reported that two participants with brainstem stroke had used the BrainGate system to control a robotic arm in three dimensions; one of participants used it to bring a thermos to her lips for a drink, guided entirely by thought.
A key practical limitation of the early system — cables connecting the brain sensor to the decoder, which tethered users to laboratory settings — was resolved in April 2021, when the BrainGate consortium reported the first human use of a high-bandwidth wireless BCI. A compact transmitter, approximately two inches across, replaced the cables and transmitted brain signals at single-neuron resolution. Two participants with spinal cord injuries used the wireless system at home to point, click, and type on a standard tablet computer, matching the performance of the earlier wired system.
Speech restoration has become a particularly active frontier. In 2023, the BrainGate2 consortium published results in Nature demonstrating that a participant with ALS, who could no longer speak intelligibly, had generated text via a BCI at a word error rate of 9.1% on a 50-word vocabulary and 23.8% on a 125,000-word vocabulary — the first successful large-vocabulary speech decoding from an intracortical BCI. In August 2024, a further BrainGate2 clinical trial participant with ALS achieved up to 97% decoding accuracy in natural, unscripted conversation and now uses the system daily to communicate with family, colleagues, and caregivers. In 2025, researchers published findings in Cell reporting that BrainGate implant users had participated in a study that successfully decoded inner speech — imagined words rather than attempted speech — from neural activity, opening a further dimension for thought-based communication.
The largest safety analysis to date, published in Neurology in January 2023, drew on 12,203 participant-days of data from 14 adults enrolled in BrainGate trials between 2004 and 2021. It identified only six serious adverse events linked to the device or its implantation, and concluded that the BrainGate Neural Interface System's safety profile is comparable to that of other chronically implanted neurological devices, including approved deep brain stimulators.
Protecting the technology: the BrainGate patent and trademark portfolio
Building an international patent portfolio was central to Cyberkinetics' strategy from the start. The company assembled a portfolio comprising patents protecting the neural interface system technology (Biological interface system), related training routine (Patient training routine for biological interface system), and a transcutaneous implant transporting signals between a sensor implanted in a patient and an external device (Transcutaneous implant), among others. To seek protection across multiple jurisdictions from a single international filing, Cyberkinetics filed a dozen applications through the PCT system, including Nerve regeneration system and lead devices associated therewith, designated in more than 100 countries.
Alongside its patent strategy, Cyberkinetics registered a portfolio of trademarks in the United States, included BrainGate™, Cyberkinetics™, NeuroPort®, Cerebus®, Bionics®, and the slogans "Turning Thoughts into Action"® and "Wired for Thought"™. In a sector where institutional credibility is critical, trademark protection served both commercial and reputational purposes.
From start-up to university stewardship: how BrainGate's IP changed hands
Despite the scientific promise of BrainGate, Cyberkinetics struggled financially. After going public in 2004, the company's stock began declining in 2006 and had fallen below one cent per share by 2007, leaving it unable to raise further capital. The company wound down and began selling its assets.
In 2009, the BrainGate IP portfolio — comprising patents, trademarks, trade secrets, and the cyberkinetics.com domain name — was acquired for just under US$1 million by Jeffrey Stibel, a Tufts University alumnus and entrepreneur. Stibel established BrainGate, Inc. to manage and advance the portfolio. At the time of acquisition, the portfolio encompassed over 30 pending and issued patents related to neural interfaces, drawing on foundational IP from Emory University, Brown University, the University of Utah, Columbia University, and MIT, as well as from the Cyberkinetics portfolio itself.
Under Stibel's stewardship, BrainGate, Inc. made the neural interface technology freely available to academic researchers and continued to develop the IP base. In September 2020, Stibel and his partners donated BrainGate, Inc. and its entire IP portfolio to Tufts University, where it now underpins the Stibel-Dennett Consortium for Brain and Cognitive Science. "Given that BrainGate's foundational technology is the basis of most, if not all, brain chips presently being developed — medical and commercial — our goal is to create a global hub for cognitive and brain science at the university that can partner with private and public entities for responsible development of this transformative technology," Stibel said at the time.
Donoghue, meanwhile, has continued as a leading figure in global BCI research, including as the founding director of the Wyss Center for Bio and Neuroengineering in Geneva, Switzerland, established in 2014.
IP as a platform for lasting neurotechnology innovation
BrainGate's IP portfolio outlasted the company that created it, sustaining a technology through ownership changes, funding cycles, and shifts in the research landscape. For innovators working at the frontier of medical technology — where clinical development timelines are long, funding is unpredictable, and commercialization is rarely straightforward — BrainGate illustrates that IP is not only a competitive tool but a form of institutional continuity. As the global BCI field continues to grow, with other entrants joining the space that BrainGate helped define, the IP framework built in the early 2000s continues to shape the development of this foundational neurotechnology.
