Fascination of investors such as Elon Musk with uploading their brains to computers is hindering progress in curing disease, say scientists<\/b>
\nFascination of investors such as Elon Musk with uploading their brains to computers is hindering progress in curing disease, say scientists
It has been an excellent year for neurotech, if you ignore the people funding it. In August, a tiny brain implant successfully decoded the inner speech of paralysis patients. In October, an eye restored sight to patients who had lost their vision.
It would just be better, say experts, if the most famous investors in the space \u2013 tech magnates such as Elon Musk and OpenAI\u2019s Sam Altman \u2013 were less interested in uploading their brains to computers or merging with AI.
\u201cIt\u2019s distorting the debate a lot,\u201d said Marcello Ienca, a professor of neuroethics at the Technical University of Munich. \u201cThere is this long-term concern regarding the narratives they use.\u201d
Michael Hendricks, a professor of neurobiology at McGill, said: \u201cRich people who are fascinated with these dumb transhumanist ideas\u201d are muddying public understanding of the potential of neurotechnology. \u201cNeuralink is doing legitimate technology development for neuroscience, and then Elon Musk comes along and starts talking about telepathy and stuff.\u201d
Silicon Valley firms have ramped up investment in neurotechnologies in the past years, with Altman in August co-founding Merge Labs, a competitor to Musk\u2019s Neuralink. Apple and Meta are both working on wearable devices that leverage neural data: a wristband for Meta, EEG headphones for Apple.
At this point, said Ienca, most of the US\u2019s big tech companies have dedicated research on neurotechnology: Google\u2019s neural mapping project, for example, or Meta\u2019s acquisition of Ctrl Labs. \u201cThe neurotech game is really in the process of going mainstream,\u201d he said.
These technologies have considerable near-term potential to treat a variety of neurological issues \u2013 from ALS to Parkinson\u2019s to paralysis. The problem is, their investors don\u2019t always appear to have curing disease as an end goal.
Musk has said brain-computer interfaces like Neuralink\u2019s may one day allow people to \u201cupload [their] memories\u201d and \u201cdownload them into a new body or a robot body\u201d. Altman, though quieter on the subject, has blogged about the impending \u201cmerge\u201d between humans and machines \u2013 which he suggested would either through genetic engineering or plugging \u201can electrode into the brain\u201d. (In 2018, Altman invested in a \u201c100% fatal\u201d brain-uploading startup, paying $10,000 to join its waiting list.)
To be clear, technologies like brain uploading are a long way off, Hendricks and Ienca said: in fact, they\u2019re probably impossible, at least in the foreseeable future. \u201cBiological systems are not like computers,\u201d Hendricks said.
Some worry, however, that far-fetched narratives could stymie actual health advances \u2013 for example, by pushing regulators to adopt broad, fear-driven laws.
Kristen Mathews, a lawyer who works on mental privacy issues at the US law firm Cooley, said all this \u201csci-fi hype could trigger regulation that would hinder advances in technology that would otherwise have the potential to really help people who need help\u201d.
\u201cIt\u2019s completely unrealistic, and it\u2019s hiding the real questions,\u201d said Herv\u00e9 Chneiweiss, a neuroscientist who chaired a panel of experts advising Unesco on its global standards for neurotechnology, adopted on Wednesday.
The actual frontier of neurotechnology is best understood as encompassing three distinct categories. There are medical devices, such as the brain implants that decode speech, or Neuralink\u2019s electronic chip that allowed a man with a spinal injury to control a computer. There are consumer wearables, a newer frontier that includes devices such as EEG earbuds or, more loosely, glasses such as Apple\u2019s VisionPro that track your eye movements.
Then there are the science-fiction efforts, such as Nectome, the brain-uploading startup, or Kernel, which aims to link brains to computers, or Neuralink\u2019s recent efforts to trademark the name Telepathy.
The first category promises the most powerful advances: restoring vision and hearing, and treating neurodegenerative diseases or perhaps psychiatric disorders. But these devices are extremely tightly regulated \u2013 as are all medical devices \u2013 and far less advanced than more hype-driven reporting sometimes suggests. A recent paper in Frontier in Human Neuroscience decried the \u201cmisleading propaganda\u201d around brain-computer interfaces, saying the technology was still in its infancy.
The second category, consumer wearables, are a thornier regulatory problem. While there has been a spate of reporting on privacy-invasive brain-measuring devices \u2013 for example, China\u2019s much-discussed EEG helmets that supposedly monitor construction workers for fatigue, or pupils for focus \u2013 it\u2019s far less clear that these have ever worked, or pose a real surveillance risk.
\u201cThe evidential robustness of the systems is very limited. There are very few replicable studies,\u201d said Ienca.
Hendricks says devices such as EEG earphones \u2013 which companies like Emotiv, for example, are now marketing \u2013 are unlikely to be an effective surveillance tool because the data is too noisy and, like signals from a lie detector, unreliable in individual cases.
Chneiweiss, however, argues that they raise real concerns: \u201cIf they are used in the workplace, they could monitor your brain fatigue or things like that, and the data could be used to discriminate.\u201d
The science-fiction applications, meanwhile, often rely on the premise that healthy people would voluntarily get invasive brain implants \u2013 to communicate with computers, for example, or move objects with their minds.
This is unlikely. If it happened, and if the technology advanced, it might well raise surveillance concerns. But, Hendricks said, it\u2019s extremely unclear if such surveillance would be significantly more useful than the vast amount of granular data \u2013 browser histories and purchase data \u2013 that big tech companies already have.
\u201cWe have so many ways to influence people through language and simple visual media,\u201d Hendricks said. \u201cI don\u2019t think that [brain implants] would catch up for a long time.\u201d
As for brain uploading, Hendricks said the idea came from people in technology who \u201cthink about computers too much\u201d, convincing themselves that the brain is hardware and the self is software that can be run on it \u2013 or in a computer, or on a robot.
\u201cIf I could really be uploaded to be made immortal in a computer, then I should be happy to just kill myself right now as long as someone tells me, like, oh, you\u2019re living in that metal box over there,\u201d he said. \u201cBut I don\u2019t think many people would take that bet. I think instinctively we know that\u2019s bullshit.\u201d<\/p>\n