Over the next few weeks, a company called Kernel will begin sending dozens of customers across the US a $50,000 helmet that can, crudely speaking, read their mind. Weighing about 1kg each, the helmets contain nests of sensors and other electronics that measure and analyse a brain’s electrical impulses and blood flow at the speed of thought, providing a window into how the organ responds to the world. The basic technology has been around for years, but it’s usually found in room-size machines that can cost millions of dollars and require patients to sit still in a clinical setting.
The promise of a leagues-more-affordable technology that anyone can wear and walk around with is, well, mind-bending. Excited researchers anticipate using the helmets to gain insight into brain ageing, mental disorders, concussions, strokes and the mechanics behind previously metaphysical experiences such as meditation and psychedelic trips. “To make progress on all the fronts that we need to as a society, we have to bring the brain online,” says Bryan Johnson, who’s spent more than five years and raised about $110m — half of it his own money — to develop the helmets.
Johnson is the CEO of Kernel, a start-up that’s trying to build and sell thousands, or even millions, of lightweight, inexpensive helmets that have the oomph and precision needed for what neuroscientists, computer scientists and electrical engineers have been trying to do for years: peer through the human skull outside university or government labs. In what must be a sort of record for rejection, 228 investors passed on Johnson’s sales pitch, and the CEO, who made a fortune from his previous company in the payments industry, almost zeroed out his bank account in 2020 to keep Kernel running. “We were two weeks away from missing payroll,” he says.
Though Kernel’s tech still has much to prove, successful demonstrations, conducted shortly before Covid-19 spilt across the globe, convinced a number of Johnson’s doubters that he has a shot at fulfilling his ambitions.
To make progress on all the fronts that we need to as a society, we have to bring the brain online
— Bryan Johnson
A core element of Johnson’s pitch is “Know thyself”, a phrase that harks back to ancient Greece, underscoring how little we’ve learnt about our head since Plato. Scientists have built all manner of tests and machines to measure our heart, blood and even DNA, but brain tests remain rare and expensive, sharply limiting our data on the organ that most defines us. “If you went to a cardiologist and they asked you how your heart feels, you would think they are crazy,” Johnson says. “You would ask them to measure your blood pressure and your cholesterol and all of that.”
The first Kernel helmets are headed to brain research institutions and, perhaps less nobly, companies that want to harness insights about how people think to shape their products. (Christof Koch, chief scientist at the Allen Institute for Brain Science in Seattle, calls Kernel’s devices “revolutionary”.) By 2030, Johnson says, he wants to bring down the price to the smartphone range and put a helmet in every American household — which starts to sound as if he’s pitching a panacea. The helmets, he says, will allow people to finally take their mental health seriously.
Johnson is something of a measurement obsessive. He’s at the forefront of what’s known as the quantified-self movement. Just about every cell in his body has been repeatedly analysed and attended to by a team of doctors, and their tests now cast him as a full decade younger than his 43 years. Along those lines, he wants to let everyone else analyse, modify and perfect their minds. No-one knows what the results will be, or even if this is a good idea, but Johnson has taken it upon himself to find out.
When Johnson founded Kernel , in 2015, his plan was to develop surgical implants that could send information back and forth between humans and computers, the way Keanu Reeves downloads kung fu into his brain in The Matrix. (In the early days, Johnson discussed a potential partnership with Elon Musk, whose company NeuraLink has put implants in pigs and monkeys, but nothing came of it.) The idea was, in part, to transfer thoughts and feelings directly from one consciousness to another, to convey emotions and ideas to other people more richly than human language allows.
Perhaps more important, Johnson believed, artificial intelligence (AI) technology was getting so powerful that for human intelligence to remain relevant, the brain’s processing power would need to keep pace.
The technology needed to make implants work is difficult to perfect — among other things, the human body tends to muddy the devices’ signals over time, or to reject them outright — and the surgery seemed unlikely to go mainstream. With the helmets, the basic principle remained the same: put tiny electrodes and sensors as close as possible to someone’s neurons, then use the electrodes to detect when neurons fire and relay that information to a computer. Watch enough of these neurons fire in enough people, and we may well begin to solve the mysteries of the brain’s fine mechanics and how ideas and memories form.
On and off for almost three years, I’ve watched as Kernel has brought its helmets into reality. During an early visit to the company’s two-storey headquarters in a residential part of Venice, in Los Angeles, I saw that Johnson’s team had converted the garage into an optics lab full of mirrors and high-end lasers. Near the entryway sat a shed-size metallic cube designed to shield its contents from electromagnetic interference. On the second floor, dozens of the world’s top neuroscientists, computer scientists and materials experts were tinkering with early versions of the helmets alongside piles of other electrical instruments. At that point the helmets looked less like 21st-century gadgets and more like something a medieval knight might wear into battle, if he had access to wires and duct tape.
Despite the calibre of his team, Johnson and his odd devices were considered toys by outsiders. “The usual Silicon Valley people and investors would not even talk to us or poke around at all,” he says. “It became clear that we would have to spend the time, and I would have to spend the money, to show people something and demonstrate it working.”
A hospital or research centre will typically employ a range of instruments to analyse brains. The list is a smorgasbord of acronyms: fMRI (functional magnetic resonance imaging), fNIRS (functional near-infrared spectroscopy), EEG (electroencephalography), MEG (magnetoencephalography), PET (positron emission tomography), and so on. These machines measure a variety of things, from electrical activity to blood flow, and they do their jobs well. They’re also enormous, expensive and not easily condensed into helmet form.
The machines’ size owes in part to components that shield the patient’s head from the cacophony of electrical interference present in the world. This allows the sensors to avoid distracting signals and capture only what’s happening in the brain. Conversely, signals from the machines need to penetrate the human skull, which happens to be well-evolved to prevent penetration. That’s part of the argument for implants: they nestle sensors right up against our neurons, where the signals come in loud and clear.
It’s unlikely a helmet will ever gain the level of information an implant can, but Kernel has striven to close the gap by shrinking its sensors and finding artful ways to block electromagnetic interference. Among its breakthroughs, Johnson’s team designed lasers and computer chips that were able to see and record more brain activity than any previous technology. Month after month, the helmet became more refined, polished and lightweight as the team made and remade dozens of prototypes. The only trick was that, to suit the different applications Johnson envisioned for the helmet, Kernel wound up needing to develop two separate devices to mimic all the main functions of more traditional machines.
One of the devices, called Flow, looks like a hi-tech bike helmet, with several brushed aluminium panels that wrap around the head and have small gaps between them. Flip it over, and you’ll see a ring of sensors inside. A wire at the back can be connected to a computer system.
This helmet measures changes in blood oxygenation levels. As parts of the brain activate and neurons fire, blood rushes in to provide oxygen. The blood also carries proteins in the form of haemoglobin, which absorbs infrared light differently when transporting oxygen. (This is why veins are blue, but we bleed red.) Flow takes advantage of this phenomenon by firing laser pulses into the brain and measuring the reflected photons to identify where a change in blood oxygenation has occurred. Critically, the device also measures how long the pulse takes to come back. The longer the trip, the deeper the photons have gone into the brain. “It’s a really nice way to distil out the photons that have gone into the brain vs ones that only hit the skull or scalp and bounced away,” says David Boas, a professor of biomechanical engineering and director of the Neurophotonics Center at Boston University.
The other Kernel helmet, Flux, measures electromagnetic activity. As neurons fire and alter their electrical potential, ions flow in and out of the cells. This process produces a magnetic field, if one that’s weak and changes its behaviour in milliseconds, making it extremely difficult to detect. Kernel’s technology can discover these fields all across the brain via tiny magnetometers, which gives it another way to see what parts of the organ light up during different activities.
0 of 2
The helmets are not only smaller than the devices they seek to replace, but they also have better bandwidth, meaning researchers will receive more data about the brain’s functions. According to the best research, the Flow device should help quantify tasks related to attention, problem-solving and emotional states, while Flux should be better suited to evaluating brain performance, learning and information flow. Perhaps the primary thing that has scientists gushing about Kernel’s machines is their mobility — patients’ ability to move around wearing them in day-to-day settings. “This unlocks a whole new universe of research,” Boas says. “What makes us human is how we interact with the world around us.” The helmets also give a picture of the whole brain, as opposed to implants, which look solely at particular areas to answer more specific questions, according to Boas.
Once their Kernel helmets arrive, Boas and his colleagues plan to observe the brains of people who’ve had strokes or suffer from diseases such as Parkinson’s. They want to watch what the brain does as individuals try to relearn how to walk and speak and cope with their conditions. The hope is that this type of research could improve therapy techniques. Instead of performing one brain scan before the therapy sessions start and another only after months of work, as is the practice today, researchers could scan the brain each day and see which exercises make the most difference.
Devices are also going out to Harvard Medical School, the University of Texas and the Institute for Advanced Consciousness Studies (a California lab focused on researching altered states) to study such things as Alzheimer’s and the effect of obesity on brain ageing, and to refine meditation techniques. Cybin, a start-up aiming to develop therapeutic mental health treatments based on psychedelics, will use the helmets to measure what happens when people trip.
All of this thrills Johnson, who continues to harbour the grandest of ambitions for Kernel. He may have given up on computer-interfacing implants, but he still wants his company to help people become something more than human.
Of course, not everyone will want to make decisions based on what a helmet says their brain activity means. Taking the decisions out of thought patterns — or analysing them for the purposes of market research and product design — poses its own, perhaps scarier, questions about the future of human agency. And that’s if the Kernel devices can fulfil the company’s broader ambitions. While the big, expensive machines in hospitals have been teaching us about the brain for decades, our understanding of our most prized organ has remained, in many ways, basic. It’s possible Kernel’s mountain of fresh data won’t be of the sort that translates into big breakthroughs. The brain researchers who are more sceptical of efforts such as Johnson’s generally argue that novel insights about how the brain works — and, eventually, big leaps in brain-machine interfaces — will require implants.
Yet scientists who have watched Kernel’s journey remark on how the company has evolved alongside Johnson, a complete outsider to the field. “Everybody he’s recruited to Kernel is amazing, and he’s been able to listen to them and motivate them,” says MIT neuroscientist Edward Boyden. “He didn’t have scientific training, but he asked really good questions.”
The test now will be to see how the company’s devices perform in the field and if they really can create a whole new market where consumers buy Flow and Flux helmets alongside their Fitbits and Oura rings. “There’s a lot of opportunity here,” Boyden says. “It’s a high-risk, high-payoff situation.”
If Johnson’s theories are correct and the Kernel devices prove to be as powerful as he hopes, he’ll be, in a sense, the first person to spark a broader sort of enlightened data awakening. He recently started a programme meant to quantify the performance of his organs to an unprecedented degree. Meanwhile, he’s taking part in several experiments with the Kernel helmets and is still looking for ways to merge AI with flesh. “We are the first generation in the history of Homo sapiens who could look out over our lifetimes and imagine evolving into an entirely novel form of conscious existence,” Johnson says. “The things I am doing can create a bridge for humans to use where our technology will become part of our self.”
Bloomberg BusinessWeek. More stories like this are available on bloomberg.com.
Would you like to comment on this article?
Sign up (it's quick and free) or sign in now.
Please read our Comment Policy before commenting.