Could Elon Musk's Neuralink brain implant become a reality?

Tech start-up's latest developments shine spotlight on emerging field of human-computer interfaces

Billionaire Elon Musk may have risen to prominence as the head of electric car company Tesla, but he arguably has grander ambitions than electrifying road transport.

Neuralink, Mr Musk's brain-chip start-up, is dedicated to forging connections between the human mind and computers, and it revealed this week that it had, in a first for the company, inserted a microchip into a person's brain.

Mr Musk reported that the subject, described as a patient but whose medical condition was not specified, was “recovering well” and that initial results indicated a “promising” interaction between the device and the person's nervous system.

In some cases, like people with Parkinson's, a chip has had significant impacts on their quality of life

Luca Citi,
professor in the School of Computer Science and Electronic Engineering at the University of Essex

Neuralink joins other companies and academic research centres that have carried out similar work as part of an emerging field of human-computer interfaces.

The focus of this work, at least in the short term, is to regain physical function in people who may have become disabled as a result of injury or neurological disease.

“The immediate or clear applications are in paralysis, injury and motor neurone disease, where people's brains are still relatively healthy but, due to injury or disease, it's no longer possible for the brain to send signals to the muscles,” Andrew Jackson, professor of neural interfaces at Newcastle University in Britain, told The National.

A chip has already helped a patient with Parkinson's disease to regain abilities thought to have been lost for ever.

In November, it was revealed that Marc Gauthier, 63, from Bordeaux in France, was able to walk almost normally again thanks to a chip that had been inserted into his spine.

“In some cases, like people with Parkinson's, it's had significant impacts on their quality of life,” said Luca Citi, a professor in the School of Computer Science and Electronic Engineering at the University of Essex in Britain.

“That's almost like the neural equivalent of a heart pacemaker or, in certain contexts, even resembling a neural bypass – a remarkable achievement in its own right.”

Prof Citi said that while it may not align with the “futuristic visions of innovators like Musk”, it signifies a “monumental leap” in neural engineering.

“When you talk about neural interfaces, however, there's often an expectation of more complex functionalities akin to smart devices,” he added.

Prof Citi’s work centres on developing ways to control upper limb prosthetics with a view to helping people to regain the ability to perform day-to-day tasks.

Key to this, he said, is being able to understand the neural or nerve signals that the person’s brain is sending, and translate them into physical movement.

However, the field is developing fast and there are many technical issues that researchers, including Prof Citi, are tackling.

Neuralink’s robotic system for inserting electrodes into the brain. Photo: Neuralink

For example, would someone who is controlling a limb by their own thoughts, but mediated through a chip, need to look at the limb to know what it is doing? Or, preferably, would they know just by feeling what the limb is doing using normal sensory feedback of the kind most of us take for granted?

“Ideally we would like to provide sensory feedback,” Prof Citi said, adding that this would resemble as closely as possible the feedback a person would receive from the nerve sensors in a normal hand.

One challenge is to train machines to “speak the same language” as sensory receptors that are transmitting impulses, so that the brain perceives these signals just as it would send signals by sensory receptors.

For implants to provide signals to the nervous system in as natural way as possible requires further development of interfaces to provide “deeper integration with biological tissue”, promoting a seamless interaction that minimises adverse bodily reactions and fosters optimal transmission of signals.

“There’s been progress over the years. There’s still a lot of research that needs to be done in this regard,” Prof Citi said.

Some of the research by Neuralink has faced criticism on animal welfare grounds, with US media reporting that several macaque monkeys with chips inserted into their brains had to be euthanised. Some were said to have suffered severe pain and distress after the chips were implanted.

Risk factors

Andrew Knight, a veterinary surgeon and professor of animal welfare at Griffith University in Australia, said work of the kind carried out by Mr Musk’s company was unlikely to lead to “tangible benefits to human patients”.

An illustration of what Neuralink's human-computer interface is to look like. Photo: Neuralink

He said surgery to implant microchips was “extremely invasive” and of high risk to the animals, including because of the likelihood of post-operative complications.

“That’s not ethical. You cannot begin to make an ethical case for a strong benefit,” he told The National.

Many people interested in human-computer interfaces are looking at the technology’s potential to augment the intellectual capabilities of the human brain, even forging links with artificial intelligence.

“As our lives become ever more closely entwined with technology – we carry a mobile phone, we spend a lot of our time online and communicating through computers – there’s a sense from some people that if we can [link] our brains directly to technology, bypassing our sensory organs and muscles, it could enhance our ability to use technology,” Prof Jackson said.

However, he described the quality of control over a computer that could be achieved through a brain implant as having “a long way to go” before it became comparable to that which could be achieved through, for example, typing on a keyboard or using a computer mouse.

So creating a situation in which a computer has become an extension of a person’s brain is unlikely in the short term.

Before scientists are able to extend what human brains can do, Prof Jackson said they would need to better understand how the functions that they want to augment – like memory or cognition – worked.

If science cannot create, for example, an artificial leg that is better than a real one, despite the human leg being well understood, how could it create a human-computer brain that is better than the original, when there remains so much more to be learnt about the brain?

Aside from technical hurdles, any new technology also has to navigate the commercial marketplace, where devices that looked good on paper can easily fall foul.

Prof Citi referred to Google Glass as an example, an “augmented reality” device that provided some of the functions of a smartphone in the form of a pair of glasses that are voice and touch controlled. They did not catch on, and were suspended early last year.

On the other hand, particular areas of technology can advance rapidly when there is great interest. “When something receives enough funding and interest, it gives a momentum and things happen faster,” Prof Citi said.