Founding and mission
Neuralink was founded in 2016 in California. Musk had already warned publicly that artificial intelligence could outpace humans. In 2017 he argued that people would have to couple their brains to technology so as not to be left “useless, or like a pet.” That existential stake still hangs over the company, even though the concrete work, for now, is medical: giving people with a spinal-cord injury, ALS, or a brainstem stroke digital independence again.
Besides Musk there were co-founders from neuroscience and engineering. One of them, Max Hodak, later left and now leads Science Corporation, which in 2026 received European approval for a vision-restoration device. Those splinters show how small and interwoven the field still is.
How the technology works
The N1 implant is about the size of a U.S. quarter and is placed in the skull. Dozens of ultrafine threads hang from it, thinner than a human hair, with electrodes that record the electrical activity of neurons — especially in areas tied to hand movement. The patient learns to steer a cursor or a command by imagining a movement. Implantation is done with a surgical robot (R1), because the threads are too fine to place reliably by hand. Fortune described in November 2026 how a robot arm drills a circle in the skull and sets the device on the brain tissue.
Telepathy is the product name Musk gave to this outbound connection: think, and a computer, phone, or game console responds. Neuralink says some participants are also testing experimental control of game consoles, keyboards, wheelchairs, and robot arms; those functions are not available to everyone in the studies. In June 2026, according to Fortune, Neuralink performed for the first time a procedure in which the dura, the protective membrane around the brain, was not opened — a step the field treats as technically important, because it can make the surgery less invasive.
Blindsight is the second product story: an implant that stimulates the visual cortex directly, so damaged eyes or optic nerves can be bypassed. In 2024 the FDA gave Blindsight Breakthrough Device status, which can speed development and review but is not market approval. Musk has said the system could even help people who have never been able to see, and that it would be ready from an engineering standpoint and waiting on regulation. Independent reporting stresses that there is still no evidence that Blindsight restores useful, usable vision in people.
Human studies and numbers
The first human implantation in the PRIME study was in January 2024, in Noland Arbaugh, who was paralyzed after a diving accident. He learned to control a cursor, played games such as Civilization VI and Mario Kart, and has later said the implant fundamentally changed his view of what he can still reach. There were setbacks too: in Arbaugh some threads retracted, which temporarily weakened the signal; software changes made up for part of that. Neuralink has made that episode partly public, after questions about transparency.
The study has since grown. Fortune wrote on November 21, 2026, that about 26 people are using the device in the test phase, compared with seven a year earlier. Musk has also cited a higher number in the same period (“more than thirty”); until Neuralink publishes a full accounting, those figures are best read as approximations. All of the named participants have serious impairments; there is no consumer product.
Financially, Neuralink remains private. Fortune cited a valuation of roughly $9.7 billion in 2025 and about $1.3 billion raised, more than double the next competitor in the field. That lead in money does not have to mean Neuralink will be first to a sales approval.
The race with Synchron and the rest of the field
Synchron, founded by neurosurgeon Thomas Oxley, places the Stentrode through a blood vessel, like a heart stent, without opening the skull. The procedure lasts, according to the company, about two hours. The signal is less rich than with Neuralink’s hundreds of electrodes, but the bar for patients can be lower. Fortune reported 15 implanted people, and Oxley said Synchron may reach FDA approval for sale sooner. In 2025 Synchron was also the first to connect to Apple’s standardized Bluetooth BCI link. A BCI, or brain-computer interface, is a system that reads brain signals and turns them into commands.
Other American players include Paradromics (speech) and Precision Neuroscience. In China, according to Fortune, the first commercially available BCI was approved in March 2026, which makes Western entrepreneurs nervous about pace and, in time, data security. In Europe, Science Corporation received approval for a vision device in July 2026 and implanted the first patient in October. UNESCO published a 2025 recommendation on the ethics of neurotechnology, signed by 194 countries, covering autonomy, freedom of thought, and mental privacy.
Recent demonstrations
On November 20, 2026, a day after the worldwide release of Grand Theft Auto VI, Neuralink posted a video of a participant playing the game with the implant, including steering a speedboat. Musk replied, “Playing GTA VI with Telepathy.” Neuralink did not name the participant and did not say which console the demo ran on. The post carried the standard disclaimer: the devices are investigational and not approved by the FDA or other regulators, and experiences of volunteer participants do not say everything about future outcomes.
Demos like that are impressive and good for awareness, but they are not clinical endpoints. They show what one participant can do in one setting, not what an average patient should expect in daily life, and not when insurers would pay for a procedure.
Controversies
Before the human studies, Neuralink came under fire over animal testing. Journalism and oversight, including from the U.S. Department of Agriculture, went to the welfare of monkeys and to whether the company tested too fast and too roughly. Neuralink denied systematic mistreatment, but the image of a Musk company that pushes “hardcore” collided here with the caution medical devices require.
There is also the gap between promise and evidence. Musk talks about curing blindness and about a future in which many more people would want an implant, not only patients with severe paralysis. Blindsight has not yet made good on that promise in people. And as soon as signals from someone’s brain go to a company server or a phone, the question arises of who may see, keep, or use that data — for care, for product improvement, or, in a more extreme scenario, for surveillance or marketing. Ethicists such as Nita Farahany call thinking and feeling “the last space of real privacy.”
What’s next
The near path is regulatory: larger studies, a frozen design, production at medical scale, and then, with luck, an FDA path for a clearly defined medical indication, probably first digital control for people with paralysis, and later vision if Blindsight makes that step. After that, insurance coverage would decide whether the technology becomes more than a remarkable experiment. The far path is Musk’s original vision: a widespread symbiosis of brain and AI. That still lies, even according to people in the field, years to decades away, and would require new laws on mental privacy. Until then, Neuralink is mainly a clinical company with a science-fiction ceiling — and one of the best-funded runners in a field that has suddenly become very crowded.