Another reason I won’t get Neuralink


I was wondering what Neuralink is good for — it must be for treating some serious medical condition, since it involves serious surgery. But no! It’s just techdude fantasies.

Neuralink’s BCI will require patients to undergo invasive brain surgery. Its system centers around the Link, a small circular implant that processes and translates neural signals. The Link is connected to a series of thin, flexible threads inserted directly into the brain tissue where they detect neural signals.

Patients with Neuralink devices will learn to control it using the Neuralink app. Patients will then be able to control external mice and keyboards through a Bluetooth connection, according to the company’s website.

An app. Bluetooth. Controlling computer mice.

It absolutely did not help that I am currently using a computer mouse, a cheap wired optical mouse, that has an intermittent fault. Every once in a while, but not often enough to motivate me to get a replacement, the LED cuts out and the buttons stop responding. The fix is to shake the cable or unplug and re-insert the USB cable. It’s a bit annoying, I really should just get a new mouse, they’re only about $7.

But now imagine that your Neuralink device has a less than perfect connection: scar tissue builds up, an electrode gets jostled out of position. Every once in a while, the app drops the Bluetooth connection. The artificial limb you’re controlling becomes unresponsive, or even worse, you miss a kill shot in Call of Duty (worse, because I’ve seen how gamers can explode in fury at the most trivial stuff). There’s no easy cable-jiggling you can do, you’re going in for major brain surgery.

Or more likely, you’ll make do as I am with my mouse…you let it slide, 99% function is good enough. The only thing is, your brain doesn’t like wires stuck in it — there will be a gradual accumulation of scar tissue and localized damage, the performance of the device will inevitably incrementally deteriorate, and Neuralink doesn’t have a good replacement strategy.

“Right to repair” acquires a new urgency when it’s a gadget imbedded in your brain. Musk doesn’t seem the type to allow outsourcing of his profitable toy, and is probably anticipating making lots of money from obsolescence.

Comments

  1. says

    80%+ of apps are spyware. Bluetooth connections are NOT secure. Major brain surgery for this dangerous folly? And then they say, ‘Patients with Neuralink devices will learn to control it using the Neuralink app’. And, while this borders on conspiracy theory territory, it is not far-fetched to think that what they leave out is that the MuskOx is probably going to hijack the mouse control and have it click on all sorts of crap, getting the neuralink victim in lots of trouble.

  2. Reginald Selkirk says

    An app. Bluetooth. Controlling computer mice.

    Which might seem like more than “techdude fantasies” if you were a quadriplegic.

  3. says

    Reinforcing what PZ said about ‘techdude fantasies: ‘ As I commented in the previous ‘nutjoblink’ post by PZ “these are likely ‘gullible ‘tech will fix everything’ idiots.”

  4. nomaduk says

    Yeah, I kind of wonder whether Stephen Hawking would have been quite so dismissive of the concept.

    Of course, one would prefer someone other than Elon Musk be in charge of it all.

  5. Matt G says

    Say, didn’t a lot of monkeys used in these experiments die? An unfortunate side effect, I’m sure….

  6. says

    I will be fascinated when hackers start literally penetrating targets’ brains. What could possibly go wrong?

    (Yes I know Iain Banks had some ideas in that vein)

  7. says

    MuskOx is probably going to hijack the mouse control and have it click on all sorts of crap, getting the neuralink victim in lots of trouble

    Did you overlook where it says you agree to literally become an unpaid twitter clickbot? Those aren’t twitches, that’s ad views baby!

  8. says

    I’m aware that this technology could be used for significant things, like controlling artificial limbs. I just don’t think it is anywhere close to being reliable or efficient for that purpose.
    I’m really curious to know how they’ll select human subjects. Getting patients with, for instance, missing limbs would be a way to emphasize the importance of the project, but would also be taking advantage of people who already have major deficits, and the failures would be more prominent. Picking individuals willing to get brain surgery so they could play pong…well, it’s more likely to actually succeed, but risks looking trivial.

  9. KG says

    Which might seem like more than “techdude fantasies” if you were a quadriplegic. – Reginald Selkirk@2

    True, but research should surely focus on ways of giving people such as quadraplegics control over their environment without brain implants, with their inevitable dangers and need for further surgery every time there’s an upgrade (the linked article unnecessarily mentions Musk, but is about the use of fMRI sensing brain activity via external contacts, and machine learning of how to interpret that activity in a specific individual.

  10. StevoR says

    If you don’t fantasise sometimes and then work to make those fantasies come true how will they ever do so?

    Not that it works for me very often if at all but still…

  11. StevoR says

    Is there anything wrong with dreaming big and trying to make it so if you work ethically and with consent and care?

    Where do we get if don’t try for ambitious dreams and fantasies and bold visionary or even “.(. hhmm.. which word t’use..) silly” ideas?

    Not syaing Iagree with this prticular one but as general principle.

    Clarke’s Laws apply. ( https://en.wikipedia.org/wiki/Clarke%27s_three_laws

  12. wzrd1 says

    So, the ever inventive and ingenious Musk has finally caught up to the 1970’s. Even if his experimental animal lab shuttered due to 1920’s level postoperative infections.
    Meanwhile, there is a brain interface that’s had partial success in restoring function after spinal cord injury abroad, which doesn’t count obviously, as Musk’s name ain’t on it.
    Next, we’ll learn of Musk’s newest inventions, a round thing that spins to help move a load that’s mounted on a stick and a lighting device that’s powered by igniting a lipophilic malleable solid hydrocarbon compound with a string in the middle. Once he figures out to light the string and not the wax…
    Which is what he gets for firing the scientific staff on the lighting project. His suggested names for the devices, Xheel and Xndle. With litigation already announced for infringement of the trademark name against every sandal company on the planet.
    I know, too believable.

    Seriously though, he’s gotten the magic interface to 1969-1970 level and I wasn’t joking about the infection rate.
    I wouldn’t trust his company to implant a fucking splinter.

  13. wzrd1 says

    Oh, I’m happy with my Kensington trackball. While it is bluetooth, if it gets hacked, who gives a shit, it’s a mouse and not a keyboard (my keyboard is wired) or block device. My system refuses to accept unencrypted block device traffic.

  14. david says

    @kg, #10 – it would have to be a very powerful assisting device to enable a user to carry around an fMRI.

    Long-term implanted electrodes are already in common use, with acceptable complication rates. Examples include cochlear implants for the hearing impaired and deep Bahrain stimulators for Parkinson’s disease and for epilepsy.

  15. Kagehi says

    @13 wzrd1

    To be even more clear, I just watched a vid as of a few days ago with someone using a similar solution, also using bluetooth (and I imagine you could add some security to them, if someone could be bothered), which “is not” based on dipshits neural link, or a flipping “app”, which already is being used to restore one person’s mobility. Its not a brain implant, I think, but connects into the spine, and the “training” is because the brain has to “relearn” how to process the signals (since obviously the original connections are likely to have been either lost, or rewired to do something else, by the brain, when connection to the parts of the body that no longer respond was lost).

    No idea of any monkeys, never mind negligently killed ones, where ever involved.

    Elon is like Google, in a way, every semi-good idea they have had doesn’t work, and everything else was someone else’s idea they stole, and/or bought from someone else. But.. he is far, far, far worse, since even the shit he does buy doesn’t work right after he has been screwing with it.

  16. says

    OMG why the FUCK did the FDA approve of human trials, when so much was known to be wrong about his animal trials? Are our civil servants that bowled over by techno-libertarian blithering?

    And no, StevoR, there’s nothing wrong with “dreaming big” — but #QElon isn’t “dreaming” big, he’s SCAMMING big, and using other people’s dreams and fantasies to sell dodgy products, which seem to have been designed (or at least imagined) with a dodgy mindset. No one can be trusted who approaches neurological problems as nothing more than “tech” problems, as #QElon’s blithe arrogant bluster strongly implies.

    A group in France (Lorach 2023 Nature) has already done this… been in use for a year.

    So let’s just go with what they’ve got, and stop acting like #QElon is the only bidnessman on Earth who’s doing anything in this field. There’s other companies in America who already have products on the market. We don’t need Musk or Theralink swanning about like they’re boldly going where none have gone before.

  17. wzrd1 says

    Kagehi, excellent knowledge on the overall subject. The rewiring is neuroplasticity, unlike the commercialized bullshit, yeah, what isn’t used gets remapped for other usage by the body.
    The monkeys had an atrocious postoperative infection rate and is currently under investigation over the test animals deaths. So, I’ve not a clue who they bought off at the FDA to get this shit approved and yes, I’m making an accusation, as such approval always would be held up pending the outcome of the existing investigation. When you’re near the water, something smells like rotting fish, one can be well assured a dead elephant isn’t in the grass.
    As for Google, their search engine was a good idea, it worked well until recently. Now, the search results are more dotty than their 0.9 version from decades ago. :/
    But, otherwise, spot on. Edison’s workshop at its best, buy everything competing, hire an assembly line of engineers and take credit for their work as your latest ingenious invention.

    Oh, on the investigation, as of Dec 2022…
    https://www.reuters.com/technology/musks-neuralink-faces-federal-probe-employee-backlash-over-animal-tests-2022-12-05/
    “In all, the company has killed about 1,500 animals, including more than 280 sheep, pigs and monkeys, following experiments since 2018, according to records reviewed by Reuters and sources with direct knowledge of the company’s animal-testing operations. The sources characterized that figure as a rough estimate because the company does not keep precise records on the number of animals tested and killed. Neuralink has also conducted research using rats and mice.”
    “The first complaints about the company’s testing involved its initial partnership with University of California, Davis, to conduct the experiments. In February, an animal rights group, the Physicians Committee for Responsible Medicine, filed a complaint with the USDA accusing the Neuralink-UC Davis project of botching surgeries that killed monkeys and publicly released its findings. The group alleged that surgeons used the wrong surgical glue twice, which led to two monkeys suffering and ultimately dying, while other monkeys had different complications from the implants.
    The company has acknowledged it killed six monkeys, on the advice of UC Davis veterinary staff, because of health problems caused by experiments. It called the issue with the glue a “complication” from the use of an “FDA-approved product.” In response to a Reuters inquiry, a UC Davis spokesperson shared a previous public statement defending its research with Neuralink and saying it followed all laws and regulations.”

    PZ covered part of the UC Davis mess here some time back.
    I don’t have a problem with animals dying after research or even for food, but I do have a problem with negligently causing suffering and wasting those animals lives needlessly. I’d have less of a problem were Musk to have been one of those animals lost…
    Read the story, you’ll likely become nauseous. I did.

    Well, off to plot my demise, erm, I mean plan dinner. Those collards aren’t gonna prepare and eat themselves!

  18. birgerjohansson says

    The cyberpunk stories had something called “Black ice”. I think I will not install a jack in my neck yet.

    As for strong AI/ general AI I am all for it… but I will be long dead when it arrives. If the rest of you inject bowhead whale DNA you might just live long enough, and start building Iain Bank’s “Culture”.

  19. hemidactylus says

    I’m looking forward to the upside of human-machine interface like when I can instantly diagnose a problem with my car as or just before it happens. Will make liar mechanics and idiot lights obsolete.

  20. wzrd1 says

    birgerjohansson, don’t be such a pessimist! I anticipate general AI to come along very soon, right around when economical fusion power comes out, which is promised to be within a handspan of years.
    Rinsed and repeated often…

    hemidactylus, I dunno, wouldn’t that just give one a slightly brighter idiot light?

  21. jaytheostrich says

    Now I don’t know much about brain surgery, but how exactly are they holding these microwires inside your squishy brain? If you bump your head hard, is your brain going to be egg-sliced? Seems like that might be a bad thing.

  22. says

    PZ covered part of the UC Davis mess here some time back.

    Yes, and one other thing he mentioned was that apparently no one was keeping sufficient records of which animals died from which causes. And insufficient documentation means that all of their animal testing, botched or not, was for nothing.

    When Theralink applied for permission to conduct human trials, the FDA should have responded with a poop emoji. That’s all that lot deserve.

  23. robro says

    I have two friends who have had “brain implants” of sorts in the past year: one to control epileptic seizures, the other to control Parkinson’s. The one for epileptic seizures was turned off after a few months, and will probably be removed eventually. It didn’t help the seizures and it caused my friends voice to go from normal to raspy constantly. I’m not sure how the one for Parkinson’s is doing, but based on a couple of conversations with my friend after the surgery he was not a miracle cure. Neither of those has anything to do with Neuralink, of course, and they are probably the result of better grounded health science research than anything Musk will allow. The fact that there are efforts in this area isn’t a surprise, nor that the outcomes are sketchy at best. Still, if someone came up with a technology and procedure that promised to eliminate…or even reduce…tinnitus, I would consider it.

  24. Anders says

    The idea is that the very thin connection threads inserted in the brain are designed so thin, and inserted so precicely they can avoid any veins etc in the brain, and , according to NL, will not be noticeable from the brain/body’s perspective, will not create scar tissue etc.

    Obviously, any surgery on the brain might be viewed as “major”, but it seems it will be pretty tiny, if the bluetooth connection drops, well.. if your a quadraplegic and have lost all connections permanently to your arms and legs, if you can move a cursor with 40% accuracy would be better than nothing. I use a bluetooth mouse, it is every bit as precise and pretty much 100% reliable. As long as it has batteries charged.

  25. mvoetmann says

    Boring. I want a built-in ALU to help with calculations. Maybe some computer vision to help me see more stuff, like more colors. Maybe a video player, so I can watch movies in my head. Radar like Daredevil. Perhaps something that will make me a better driver. And increase my juggling skills.
    Until they can deliver some of that, never mind.