Uploading the Human Mind Could One Day Become a Reality, Predicts Neuroscientist (sciencealert.com)
- Reference: 0177894355
- News link: https://science.slashdot.org/story/25/06/01/1926231/uploading-the-human-mind-could-one-day-become-a-reality-predicts-neuroscientist
- Source link: https://www.sciencealert.com/uploading-the-human-mind-could-become-a-reality-expert-says
"But as of today, [3]we're [4]nowhere close ..."
> Replicating all that complexity will be extraordinarily difficult. One requirement: The uploaded brain needs the same inputs it always had. In other words, the external world must be available to it. Even cloistered inside a computer, you would still need a simulation of your senses, a reproduction of the ability to see, hear, smell, touch, feel — as well as move, blink, detect your heart rate, set your circadian rhythm and do thousands of other things... For now, researchers don't have the computing power, much less the scientific knowledge, to perform such simulations.
>
> The first task for a successful mind upload: Scanning, then mapping the complete 3D structure of the human brain. This requires the equivalent of an [5]extraordinarily sophisticated MRI machine that could detail the brain in an advanced way. At the moment, scientists are only at the very early stages of brain mapping — which includes the [6]entire brain of a fly and [7]tiny portions of a mouse brain . In a few decades, a complete map of the human brain [8]may be possible . Yet even capturing the identities of all [9]86 billion neurons , all smaller than a pinhead, plus their [10]trillions of connections , still isn't enough. Uploading this information by itself into a computer won't accomplish much. That's because each neuron [11]constantly adjusts its functioning , and that has to be modeled, too. It's hard to know how many levels down researchers must go [12]to make the simulated brain work . Is it enough to stop at the molecular level? Right now, no one knows.
>
> Knowing [13]how the brain computes things might provide a shortcut. That would let researchers simulate only the essential parts of the brain, and not all biological idiosyncrasies. Here's another way: Replace the 86 billion real neurons with artificial ones, one at a time. That approach would make mind uploading much easier. Right now, though, scientists can't replace even a single real neuron with an artificial one. But keep in mind the pace of technology [14]is accelerating exponentially . It's reasonable to expect spectacular improvements in computing power and artificial intelligence in the coming decades.
>
> One other thing is certain: Mind uploading will certainly have no problem finding funding. Many billionaires appear [15]glad to part with lots of their money for a shot at living forever. Although the challenges are enormous and the path forward uncertain, I believe that one day, mind uploading will be a reality.
"The most optimistic forecasts pinpoint [16]the year 2045 , only 20 years from now. Others say the end of this century.
"But in my mind, both of these predictions are probably too optimistic. I would be shocked if mind uploading works in the next 100 years.
"But it might happen in 200..."
[1] https://www.sciencealert.com/uploading-the-human-mind-could-become-a-reality-expert-says
[2] https://scholar.google.com/citations?user=tAaE5jIAAAAJ&hl=en
[3] https://alleninstitute.org/news/why-is-the-human-brain-so-difficult-to-understand-we-asked-4-neuroscientists/
[4] https://youtu.be/2DWnvx1NYUA
[5] https://magazine.columbia.edu/article/machine-can-read-your-mind
[6] https://www.bbc.com/news/articles/c0lw0nxw71po
[7] https://apnews.com/article/brain-map-neurons-alzheimers-autism-1a4e9db0a86c082e10da9c154546c592
[8] https://www.technologyreview.com/2018/03/13/144721/a-startup-is-pitching-a-mind-uploading-service-that-is-100-percent-fatal/
[9] https://www.nature.com/scitable/blog/brain-metrics/are_there_really_as_many/
[10] https://hms.harvard.edu/news/new-field-neuroscience-aims-map-connections-brain
[11] https://www.verywellmind.com/what-is-brain-plasticity-2794886
[12] https://youtu.be/XmQOW9udaQ0
[13] https://magazine.columbia.edu/article/your-beautiful-brain
[14] https://singularityhub.com/2016/03/22/technology-feels-like-its-accelerating-because-it-actually-is/
[15] https://www.marketwatch.com/story/upload-your-mind-or-alter-genetics-powerful-billionaires-are-pouring-money-into-life-extending-technology-and-they-just-might-succeed-6e1042f4
[16] https://bigthink.com/the-well/will-humanity-become-digitally-immortal/
How to write a clickbait story (Score:3)
Step 1) Pick something from a movie/story that is fantastical.
Step 2) Find someone with a degree that is either unethical or stupid enough to claim that it will be done 'in the future'.
Step 3) Have the reporter pick a time in the future that seems reasonable to a layman.
Step 4) Pretend you did not do any of the earlier steps.
Step 5) PROFIT!
Their description of 'uploading' says "replicating", which = a copy. If you copy a human mind, you are not being uploaded, you remain in your human body. At best they have cloned your mind into a robot. If they kill the original you, you still die - even if they do it just after the 'upload'. Do not let them copy you then murder you , even if the copy hides the fact that they murdered you. (AKA the Star Trek Transporter Problem).
If you want to move your consciousness to a computer, you need a slow and steady partial replacement of bioware with hardware. Think "Ship of Theseus" methodology with long time periods - only replacing the organic parts with inorganic parts.
That would actually let you upload. But this technology does not exist in any way shape or form. Neither the hardware nor a process to meld them with our existing bioware.
Re: (Score:2)
I definitely agree with the first part of what you said. There's an awful lot of hand-waving going on here. Trying to put a timeline on things that have yet to be discovered or invented is dumb, no matter what the credentials of the person making the claims.
Re: (Score:2)
> Their description of 'uploading' says "replicating", which = a copy. If you copy a human mind, you are not being uploaded, you remain in your human body.
Every upload of anything is a copy. If your mind were uploaded, then, yes, the original "organic" you would remain in your body. But if the machine to which the copy were uploaded to were able to "run" your mind (a big "if," granted), then there would be two of "you."
Re: (Score:2)
I will explain the "Ship of Theseus" method of uploading.
Part 1) Ship of Theseus: Ship gets damaged over time, carpenter slowly replaces each damaged part. After X amount of time, the carpenter realizes he has replaced every single piece of wood in the ship. Questions: Is the current ship still the original? When did it change? If there was never damage, just weathering and the carpenter kept all original pieces and re-assembeled it, which would be the actual ship of Theseus?
Part 2) How to actually up
Re: (Score:2)
I like this method a lot.
Re: (Score:2)
> If you want to move your consciousness to a computer, you need a slow and steady partial replacement of bioware with hardware.
I'd like to recommend the very thought-provoking short story [1]Learning To Be Me [wikipedia.org], written by Australian writer Greg Egan (if you don't want spoilers, the Wikipedia article contains a plot summary). It's quite topical to this discussion, and also quite unsettling.
[1] https://en.wikipedia.org/wiki/Learning_to_Be_Me
Re: (Score:2)
I'd recommend the game [1]SOMA [wikipedia.org]. It has nothing to do with Brave New World, and features brain uploading quite heavily. It's from the studio that made the Amnesia series.
[1] https://en.wikipedia.org/wiki/Soma_(video_game)
Re: (Score:2)
> If you copy a human mind, you are not being uploaded, you remain in your human body.
You'd think someone posting on Slashdot would have done more uploading. Maybe you're more familiar with downloading? The original tends to remain. Otherwise things like Netflix would be *very* expensive.
I wouldn't bet it'll take a long time (Score:1)
Look at where things have gone in the past 50 years. Much of the stuff we interact with on a routine basis today was science fiction in 1975.
Internet. Cell phones/ Computer that actually talk and respond to voice commands. Self driving cars. All of the world's collected knowledge available in the palm of your hand, from pretty much anywhere you are.
Artificial intelligence was only in apocalyptic movies until Chatgpt was announced (and, lets not kid anyone here), stunning the world with it capabilities
Re: I wouldn't bet it'll take a long time (Score:2)
Oh, AI has been around a lot longer than ChatGTP. But ChatGTP represented an inflection point where the LLM technology was abruptly recognized as more important than it had been before. All of a sudden everyone's research projects graduated into full products.
Re: (Score:2)
"Full Products" is a misnomer.
They believed that certain models would make money, and attempted to monetize them, which has been mildly successful.
There are certain models with great degrees of accuracy, but aren't really designed as LLMs. Most LLMs are embryonic and are fraught, despite what the marketing prattle will tell you.
There are a few LLMs that can accurately solve small problems consistently, and a few with what appears to be creativity.
LLMs as a whole, however, can solve a few tasks with confiden
Re: I wouldn't bet it'll take a long time (Score:2)
It's no longer a "more data" problem, it's now a "more layers" problem. But still important progress.
Re: (Score:2)
Progress, yes. Model collapse, yes. A plateau emerges.
Re: I wouldn't bet it'll take a long time (Score:2)
Our brains work. Therefore, there are methods to take this further. We just haven't found them yet.
Re: I wouldn't bet it'll take a long time (Score:1)
Our tech is impressive compared to the 1970's but it's still doing more or less the same functions, just faster and cheaper. AI is a very optimistic description, we're not anywhere close to human intelligence, just reproduced some of its fearures. That doesn't dismiss LLM as a useful cognitive tool, but it is still just a tool, as useless as a hammer without a human operator.
Re: (Score:2)
The voltage clamp was invented in 1947, the microscope in sixteenth century and programmable computers, well, a long time ago. With those elements plus enough "faster and cheaper" you've got brain uploading.
You need a lot of faster and cheaper though.
upload them to war bots that can be on real battle (Score:2)
upload them to war bots that can be on real battle battlefield.
Shut the F up (Score:1, Insightful)
Intelligence is not just series of computations and 0s and 1s. This is fantasy on the level of 'one day i can move things with my mind'
Re: (Score:2, Insightful)
As a thought experiment, suppose we coded a simulation that builds a model of a planet's gravitational field based on its mass and shape. Then, we input very detailed data of the mass and shape of the planet earth, and it builds a model of the earth's gravitational field in its memory.
Have we just "uploaded the earth's gravitational field?" No, and the question is clearly silly.
Similarly, building a simulation of a brain isn't the same as "uploading one's mind." The very concept of "uploading the mind" i
While there are certainly questions (Score:1)
About whether a copy would be you or not I think it's safe to say that the human brain is a physical thing that can be copied given enough capability and technology.
There is some question about the Fidelity of the copy of course since if you start getting into quantum physics at some point things might go off the rails.
But overall I think assuming our civilization doesn't collapse, which I think I'm on record saying is assuming probably a bit too much but, if we keep advancing then yeah eventually w
Re: (Score:2)
> Intelligence is not just series of computations and 0s and 1s. This is fantasy on the level of 'one day i can move things with my mind'
It's a clear scientific statement. There are plenty of people who have made this assertion. Now prove it. Nobody who made it before actually has. Searle's Chinese Room is pure sophistry.
Re: (Score:2)
Searle's Chinese Room is misunderstood, especially today.
The essential feature of the Chinese Room is that the translation book doesn't change. Searle sort of associated that with "programs" and AI, which was fairly reasonable to do in the 1980s when the dominant paradigm for AI involved compiling a massive database of facts and applying the right rules to manipulate those facts. Under such restrictions it's reasonable to assume that "true" intelligence is impossible.
Real computers have access to modifiable
Fucking Bullshit (Score:5, Insightful)
We've had the full wiring diagram of C. elegans -- all 302 -- for decades. Still can't predict its behavior. But hey, who needs details when you're uploading 86 billion human neurons? Neural lace, brain emulation, consciousness on a chip -- so close, just a few Nobel Prizes away.
You could argue that we don't need to understand the brain in order to emulate it. That may be true, but does that mean we have to emulate neurotransmitter states? Dendritic potentials? Local field dynamics? Epigenetic states? Membrane lipid configurations? Where do we draw the line? If we go deep enough, not even all the compute power in the world would be enough to emulate a single human brain in real time.
The human brain is the most complex known structure in the observable universe. It is probably more complex than the underlying physics and the grand unified theory of this world, if such a theory exists.
Re: (Score:1)
Please try posting something NOT related to an election or politics. Especially when it's grossly off-topic. Thanks.
Orange Catholic Bible (Score:2)
Thou shalt not make a machine in the likeness of a human mind.
Thou shalt not disfigure the soul.
Prepare to be surprised (Score:2)
Something I read today, from a review in Nature of a book on Space Rockets by Hermann Oberth:
"A voyage to the moon would be an atractive trip to many adventurous spirits, and in these days of unprecedented achievements one cannot venture to suggest that even Herr Oberth's ambitious scheme may not be realized before the human race is extinct."
That was written in 1924. 45 years later, man was on the moon.
Re: (Score:2)
Similarly, in October 1903, an editorial in the New York Times claimed it would take between one and ten million years for humans to develop heavier-than-air machines that could fly. The Wright Brothers proved that wrong just under ten weeks later.
[1]https://en.wikipedia.org/wiki/... [wikipedia.org]
[1] https://en.wikipedia.org/wiki/Flying_Machines_Which_Do_Not_Fly
Hilbert vs Goedel (Score:2)
Is it reality or does it just describe reality?
I'd Prefer Mind Downloading (Score:3)
I'd prefer mind downloading. That is the ability to directly download into my brain knowledge, perhaps even skills.
Some might say that we already have this. Technically watching a YouTube video could qualify for this. But it's far far slower than I'd like. I want to spend a few hours downloading Wikipedia into my brain so that I can conquer my world tomorrow.
Re: (Score:2)
In other words, you want to be Neo.
Re: (Score:2)
If there's no spoon, how do we eat?
Hollywood did it 42 years ago (Score:3)
Natalie Wood's last movie: [1]Brainstorm [youtu.be]
[1] https://youtu.be/ueYcVmH0edk?si=16qKjVV3dtkjBIZQ
Somebody read Fall (Score:2)
;or, Dodge in Hell (Neal Stephenson)
uh no (Score:2)
> I fully expect mind uploading to one day be a reality.
The mind is made up of both the wetware and its state. Making a copy of both means both making a physical copy of the brain, and also somehow copying state from the source brain to the target brain.
If you built a machine and uploaded the same state to it, it still wouldn't be a brain.
Cryogenics (Score:2)
That's why I don't buy the argument that cryogenics is a con because neurons get too damaged to ever work again. Just scan and simulate.
So the wealthy don't need scientists to get a move on. But they do risk being revived into a "I Have No Mouth, and I Must Scream" dystopia.
Memory requirements (Score:2)
640k ought to be enough for anyone
Grow a Brain (Score:2)
Why not grow a new brain and body. I'd wager a decent AI could assist with a business plan. This almost sounds like a lot of movies.
Even 200 years seems optimistic (Score:2)
We famously don't know what we don't know when it comes to the human brain, so any prediction is more a thing of fairy tales than anything actually based in reality.
Re: Even 200 years seems optimistic (Score:2)
Haha... so true ... I think were seeing this all around these days... whoda thunk quantum physics would take so long huh? The next big thing might be harder to do than a snap of the fingers!
Take it step by step. (Score:2)
You don't need to simulate all that, at least initially. Scan in the brains of people who are at extremely high risk of stroke or other brain damage. If one of them suffers a lethal stroke, but their body is otherwise fine, you HAVE a full set of senses. You just need to install a way of multiplexing/demultiplexing the data from those senses and muscles, and have a radio feed - WiFi 7 should have adequate capacity.
Yes, this is very scifiish, but at this point, so is scanning in a whole brain. If you have th
It's a copy, not a move (Score:2)
If you have your brain scanned and emulated on a zsbrain, you and the zsbrain will exist at the same time.
So you don't "get to live forever", the zsbrain does.
Accelerating exponentially (Score:2)
The maths pedant in me would like to point out that every exponential function also accelerates exponentially. So the “accelerates” is redundant and “grows exponentially” is sufficient.
As you were.
Re: (Score:2)
I prefer to state that the jerk increases exponentially. Since every derivative is exponential, you can just pick your favourite.