Musk aims to get inside all of our heads with his brain chip
It may sound more like science fiction but the billionaire space and car chief will unveil his neural lace prototype,
In a science fiction series known as The Culture, Scottish novelist Iain M Banks dreams up a future, interstellar society in which humans made advancements in intelligence with a brain-computer interface known as “neural lace” implanted in their skulls.
For Elon Musk, the billionaire chief executive of Tesla and SpaceX and self-admitted fan of neural lace, taking a crack at his own version seemed a no-brainer, leading to the launch of a secretive firm in 2016 called Neuralink that would explore the technology’s potential.
Four years on, and Musk is ready to show the world his plans to hack the human brain. The entrepreneur will today unveil a demonstration of a prototype device that could one day be placed inside our skulls, making us part-human, partcyborg.
His reasons to do this are two-fold. At one level, he sees an implanted medical chip working as a brain-computer interface that could help us better understand neurological diseases and treat them.
At a deeper level, the chip is meant to boost human brain power so that we can keep up with a future shaped by artificial general intelligence. Even in a “benign AI scenario” we will be left behind, Musk has previously asserted. “If you can’t beat it, join it,” he once claimed.
But how exactly would it work, can it live up to the hype, and what would it be used for? Though details have been relatively scarce, there is some insight into the mechanics behind the Neuralink chip. At four by four millimetres, it is claimed the chip will connect to a thousand “threads” thinner than a human hair, which enter the brain through four holes drilled into the skull. Electrodes embedded in the threads, which are “sewed” to the chip, can then pick up electrical signals from neurons that indicate activity in the brain that is then translated into movement or behaviour.
In the medical arena, early versions of the implant could allow paralysed patients to control their smartphones. Later versions could build on existing procedures such as deep brain simulation for the treatment of Parkinson’s disease.
Alongside any therapeutic applications, Musk’s goal is to advance human intelligence with the chip. He sees it sitting on top of the limbic system, the more primitive part of the brain, and the cortex, the cerebral part, as a third layer that augments their intelligence. Some scientists are bullish on the prospects. Prof Paul Matthews, head of the department of brain sciences at Imperial College London, believes the challenges in understanding brain signals are more an engineering problem, which Musk is focused on overcoming.
“If you have a very thick electrode and you’re recording from hundreds of nerve cells, understanding how they’re communicating among themselves is a challenge,” he says. “But if you can get down to one or just a few, it becomes easier to decode.”
There is some precedent for what Musk is attempting to achieve. BrainGate, an organisation formed of a team of researchers from top universities in the US, has kicked-off clinical trials for a brain-computer interface with a similar ambition.
By implanting a chip in the brain, as proposed by Musk, BrainGate’s aim is to use technology to restore communication and mobility in people with neurological disorders, such as motor neurone disease.
In one trial, the micro-electrodes of the chip and a computer managed to “decode” neural signals in real-time of people with paralysis in their hand and arm, allowing them to control a cursor on a computer “simply by thinking about the movement”.
“The technology exists today in more than just a proof of concept for
recording from many neutrons in the brain, decoding that to understand something about the way a limited section of the brain is making computations,” Matthews says.
Another alternative to Neuralink gaining traction is built by a Californian firm called Kernel, which believes its non-invasive technology that allows a user to simply pop a helmet on and off that could work as a Fitbit for the brain.
The company told The Daily
Telegraph that its helmet could, for example, assess the impact of an overconsumption of news on the brain, allowing people to curate a healthier information diet through a downsized, more personalised version of brain scanning machines currently used in neuroscience laboratories and hospitals.
Grabbing data from brain scans in this way could open the doors to any number of applications, as information about the brain could become monetised in the way that monitoring our heart rate, blood pressure and sugar levels have in other areas.
A history of trials in the field makes the Neuralink vision of the future more than a fanciful vanity project then, but not everyone is convinced of its ability to deliver.
Prof David Curtis, of UCL’s biosciences division, sees some limited applications from the kind of deep brain simulation the Neuralink might offer, particularly in treating things such as obsessive compulsive disorder. But the use of it to advance intelligence by an order of magnitudes? That’s much more complicated. Curtis explains that to really achieve this, any sort of technology would really need to be able to parse out the activity of every single individual neuron in the brain and trace its connections.
Though electrodes are getting ever closer to this, the challenge of putting this application into practice is still a monumental one. “Being able to produce thoughts that drive phones and drive machines, this is the impression this work is creating and this is the bit that is fanciful,” he says.
Earlier this month, Musk claimed the demonstration would reveal “neurons firing in real time. The matrix in the matrix” and added “it will blow your mind haha” on Twitter.
While Musk will reveal just what this means at 11pm tonight, some are sceptical. A report in science magazine
Stat claimed former employees were concerned about Neuralink’s aggressive timeline. One account claimed researchers attempted to wire thousands of electrodes into a sheep’s brain in a single surgical procedure, which “failed”. Neuralink said some claims in the piece were inaccurate.
It is not just technological barriers in the way. The regulatory hurdles ahead are likely to stretch out the timespan in which any feasible product could be commercialised. Musk admitted in 2019 that all of this will be a “slow process”, with a stamp of approval needed from the US Food and Drug Administration.
“All of this will occur quite slowly. It’s not going to be like suddenly, Neuralink will have this incredible neural lace and start taking over people’s brains. It will take a long time, and you’ll see it coming,” he said.
Meanwhile, a report published yesterday by RAND, a non-profit think tank, detailed how such technology was gaining increasing attention from the US military in combat scenarios, and would warrant close scrutiny at a legal and ethical level.
Despite the hurdles, in Musk’s hands it is unwise to rule anything out.
Elon Musk will unveil his prototype chip in the brain later today. The tycoon hopes it could help advance human intelligence