Synchron is pushing its brain-computer interface beyond typing and screen control, showing how AI, spatial computing, and an implanted device could work together inside a home. In a demonstration at the Nvidia GTC conference this week in San Jose, California, the company showed trial participant Rodney Gorham controlling connected devices from his living room in Melbourne, Australia.
From Typing to Controlling a Home
Gorham is paralyzed and has lost the use of his voice and much of his body due to amyotrophic lateral sclerosis, or ALS. The degenerative disease weakens muscles over time and eventually leads to paralysis. He received Synchron’s implantable brain-computer interface, or BCI, in 2020.
At first, Gorham could use the BCI to type on a computer, iPhone, and iPad. The newer system expands that idea into the physical environment. Using the Apple Vision Pro, he can look at devices around his home and see a drop-down menu overlaid on what is in front of him.
With the BCI, he can select actions by thinking. In the demonstration, Synchron showed him playing music from a smart speaker, adjusting lighting, turning on a fan, activating an automatic pet feeder, running a robotic vacuum, and adjusting the temperature on his air-conditioning unit.
That shift matters because many BCI demonstrations have focused on single tasks, such as playing a video game, moving a robotic arm, or piloting a drone. Synchron is aiming at something broader: a system that can handle many practical actions in the home, where context changes constantly.
Why Nvidia’s AI Is Part of the System
Brain-computer interfaces work by decoding signals from brain activity and translating them into commands on an output device. Synchron is using Nvidia’s Holoscan, an AI sensor-processing platform, to improve the speed and accuracy of that decoding.
For a BCI user, speed and accuracy are not abstract performance goals. Faster and more accurate decoding could shorten the delay between the user’s intended movement and the system’s command. It could also make control more precise.
Tom Oxley, Synchron’s CEO, told WIRED that the goal is real-world operation rather than a controlled lab trick. “It’s running in real time, in a real environment 24/7, making predictions where context really matters,” he said.
That is the core challenge for a smart home BCI. A person may be looking at one device, choosing between multiple options, and expecting the system to understand the intended action without a long pause or repeated corrections. The more devices and choices involved, the more important reliable decoding becomes.
The Push Toward Cognitive AI
As part of its collaboration with Nvidia, Synchron is developing what Oxley has called “cognitive AI.” The term refers to combining large amounts of brain data with advanced computing to create more intuitive BCI systems.
Oxley sees cognitive AI as the next phase of AI development after agentic AI, which can act and make decisions independently, and physical AI, which brings AI into robots and other physical systems. In this vision, the BCI becomes less like a narrow input device and more like a flexible control layer for the world around the user.
David Niewolny, senior director of health care and medtech at Nvidia, described the demonstration as an early step. “What we saw Rodney do is a start, but there are so many more interactions that you can actually begin bringing here,” he said. With cognitive AI, he said, the mind will be the “ultimate user interface.”
Today, BCIs are usually trained with data from a single person. A user may be asked to perform a task such as thinking about moving a cursor left or right. An electrode array collects neural activity while that task is happening, and researchers label the brain data so an AI model can learn which pattern corresponds to which intended movement.
Synchron plans to use brain data from current and future trial participants to build an AI model. Maryam Shanechi, a BCI researcher at the University of Southern California and founding director of its Center for Neurotechnology, said a brain foundation model could improve accuracy and support a wider range of functions without requiring hours of training data from each individual patient.
“This model would be more generalizable, more accurate, and then you can fine-tune it in each subject,” she says. “Because this AI has been trained on the brains of many people, it has essentially learned how to learn, how to think, and then you have this brainlike AI system that you can use for a variety of tasks.”
Training Still Matters
Even with a broader model, each new BCI user would still need some training. Users learn to operate a BCI through prompts such as “squeeze your fist” or “press down like a brake pedal.” A paralyzed person may not be able to make that movement, but neurons in the motor cortex can still fire when they attempt it.
Those intended movement signals are what the BCI decodes. Oxley said Synchron will use Cosmos, Nvidia’s new family of AI models, to generate photorealistic simulations of the user’s body. The idea is that the user can watch an avatar of their own movement and mentally rehearse it.
Cosmos can also generate tokens about each avatar movement that work like time stamps. Those tokens will be used to label brain data, helping an AI model interpret and decode brain signals and then translate them into the intended action.
All of that data would contribute to a brain foundation model, described as a large deep-learning neural network that can be adapted to many uses instead of being trained from scratch on each new task. Shanechi noted a central limitation: foundation models need a lot of data to become truly foundational, and that is difficult for invasive technology that only a small number of people will receive.
A Less Invasive Path, but Still an Ambitious One
Synchron’s implant differs from some competing approaches. Neuralink and other companies use electrode arrays that sit in the brain or on the brain’s surface. Synchron’s array is a mesh tube inserted at the base of the neck and threaded through a vein to read activity from the motor cortex.
The procedure is similar to implanting a heart stent in an artery and does not require brain surgery. Vinod Khosla, founder of Khosla Ventures, one of Synchron’s investors, said that familiarity could matter for scale: “The big advantage here is that we know how to do stents in the millions around the globe. In every part of the world, there’s enough talent to go do stents. A normal cath lab can do this. So it’s a scalable procedure.”
As many as 2 million people in the United States alone receive stents every year to prop open their coronary arteries to prevent heart disease. Synchron has surgically implanted its BCI in 10 subjects since 2019 and has collected several years’ worth of brain data from those people.
The company is preparing to launch a larger clinical trial needed to seek commercial approval of its device. The field has not yet seen large-scale trials of implanted BCIs because of the risks of brain surgery and the cost and complexity of the technology.
Synchron’s goal remains ambitious and carries risks. Nita Farahany, a professor of law and philosophy at Duke University who has written extensively about the ethics of BCIs, said the more immediate possibility is “more control over more in the environment.” For people with paralysis, that is already a meaningful target: not just using a computer, but reaching into the home through thought-driven control.