@futurebird I liked the depiction in @pluralistic’s Walkaways - the simulation of your brain has look-ahead and has to forcibly keep you from experiencing existential crisis over being a computerized simulation of yourself.
@futurebird here's the mandatory reference to The Emperor's New Mind https://en.m.wikipedia.org/wiki/The_Emperor's_New_Mind by Roger Penrose who claims that there's something ineffable about the human mind that can never be eff'd. Specifically he argues that mind and/or consciousness are emergent properties of the quantum nature of the underlying biological systems, and, therefore cannot be modeled by any algorithmic system.
Yeah... I don't know about all that mystical stuff. I don't think it's necessary. I sincerely think that if we study the brain, and stop thinking of it as a computer and our memories as binary data and the "soul" as a mystical glowing orb... we'll figure out why consciousness exists.
It may simply exist because it must, having an "agent" to juggle choices and predict the future is a good adaptation. (But I do think we want to understand it more than that...)
@futurebird@davidbrin I agree. In the last decade there had been tremendous progress in understanding off both mind and consciousness, but not from psychology. Instead, it has been neurology that has made inroads into an area that wasn't formerly part of their milieu... not that I cream conjure a single reference asst the moment 😏
Consciousness exists as a delivery platform for feelings;
Feelings (Hunger, desire, fear, etc) exist as metrics of need;
Needs only exist pursuant to a persistence/survival imperative (i.e., it doesn’t matter if you’re about to starve unless you want to stay alive).
If true, then providing a virtual machine on which to run consciousness would need to include some simulacra of those needs.
I've always liked the idea that part of the function of consciousness is to rewrite the lower levels of brain function to meet the needs of the current environment (ex. I don't remember the drive home, because nothing novel happened and my consciousness never had to engage).
@futurebird@davidbrin@futurebird boosted grandparent and adding a mention of @gregeganSF who has written a lot of brilliant Sci-fi on the topic of uploading. I suspect Mr. Egan might also be sympathetic to Roger Penrose's quantum arguments since many of his stories feature a qubit for a truly conscious mind, hence my reply to my own part on the topic.
I think it will be doable in the next 25 years or so, but that doesn't mean it will be possible, or, for a lack of a better word, compatible.
A lot of what defines our mind state maps to neurons, muscles, spatial position, synesthesia. Someone waking up on a bio-mechanical body will suffer a huge shock, and I don't think it will survive or remain sane after that initial confusion. Will be a cruel torture.
Theoretically, it should be possible to simulate absolutely anything. I see no fundamental reason why a sufficiently advanced computer shouldn't be able to correctly simulate a complete brain.
But the practical problems are staggering. We're barely scratching the surface on how the human brain actually works. We don't even know what we mean by the word "intelligence". Not to mention the question of what consciousness is. And we somehow want to read and map billions of neurons and their trillions of connections, in a non-invasive way that preserves every nuance?
I'm not going to say it's impossible, but the practical obstacles are staggering. They could certainly be unsurmountable.
@futurebird I went with probably possible, but want to clarify that "human mind" is another one of those things like "species" where it doesn't correspond to any "natural" category in the first place.
Uploaded you couldn't possibly prove it's a successfully uploaded version of the same "self" as original you anyway. But it seems like it's theoretically possible for sentience and qualia to emerge on mechanical substrates, so that in principle an uploaded you could "experience" being you.
@futurebird
I think it will be but don't expect it to be possible with the current style computers. I feel like making it fully digital is not going to work (can't scale enough), need some "analogue" processing and storage too
@futurebird
Ever is a long long long long long #time. It only took 4 B years to go from a dead dry airless rock to a sentient space faring species. And that whole space faring #anthropocene era has only been 100/4B = 0.0000025 % of that #planet's life. Hard to imagine 40,000,000 more centuries of inventions like the #computer, #Internet, space shuttle, CRISPR, #nuclear reactors, #AI, #quantum computers, and space shuttle...
@futurebird "When a distinguished but elderly scientist states that something is possible, he is almost certainly right. When he states that something is impossible, he is very probably wrong." — Arthur C. Clarke
@futurebird The concept of “self” is the toughy, I think. If we could instantaneously duplicate every atom in my body 3 ft to my left, I’m quite certain the duplicate won’t be me. He probably will have his own “self”.
So how do we upload and preserve “self”. Because, as far as I’m concerned, there’s no way I’ll be what gets uploaded.
Exception would be an incremental process where “self” is preserved at each increment.
@futurebird A static brain could be a copy of a static brain, but brains are never static. You'd notice if you suddenly had better memory (or worse), if how you felt after smoking weed and listening to music was different.
I once tried a small dose of Prozac and it was one of the worst days of my life. A reconstituted brain could be so much worse.
@futurebird “Ever” could be a very long time. We have everyday things now that are pure magic from just 50 years ago, so 1,000 or 10,000 years from now, you bet.
@futurebird I would say it's hard to say either way. It really does depend on how everything works.
I think if the mind is separate from the body (and I do see it as such, as empirically proven in albeit small samples - multiple "minds" can inhabit one body, the "mind" is not inherently limited/defined by the body, etc) it's technically possible - it's just a matter of understanding how to emulate the conscious mind as its own thing, independent of its embodiment.
Realistically? I still think it's impossible, and I think almost all the "research" being done toward it right now (Musk's crap, etc) is worthless at best and a human rights violation at worst.
That said, that "research" and development doesn't invalidate the possibility that someday, a few hundred years from now, if we survive, we'll figure out how to make the individual mind/consciousness portable/capable of being "backed up"
@futurebird I'm agnostic, on one hand, a thousand years from now we could have biological computers working on the quantum level with untold capabilities, on the other and far more likely hand a thousand years from now the most advanced technology has been a rock for nine centuries(forget clubs you'd need trees for those). As for the current computer paradigm I'm a hard no. That shit will never have a thought.
@Skovheks LOL yeah, "space visitors" aka astronauts. Often complain: on one hand they'd get horny af in space because of gravity and just being human. Assuming they figure out a way to commence docking procedures. The males parts would veer wildly between stiff and flexable. but womens not so much. And similarly the assembly process for a new kid...well stuff goes sideways pretty quickly from mutated workers on the way to mums factory (eggs, and etc) ok so now kiddo is in the building phase now what: well prepare for one of the weirdest looking most mal adjusted humans ever. I think NASA and chinase equivilant fairly well now think they'd have potential for be a smart effer though: turns out rotating as fast as the earth (weightlessness). meens the head could (theortically) be comfortable enormous same for the brain etc. the rest of the body would, well like you said: go to utter ass though.
@Skovheks When I was a yute I loved the living crap out of jupitor. I'd read book after book on the topic. Well when I was in gradeschool, we had to make a science project that, if built, would work. I naturally decided to make a model of a workable place to live on jupitor. And what a native alien might be like their. What I came up with for a native ended up looking like a cool look pretty large jellfish+elf like humanoid. And guessed (theorized) : as long as they stayed close to the upper bits in very compact, floating homes and cities. they could live their just fine. I think I even devised, a travel system that'd use very compressed gas to go flying around place to place.
lol now many years latter we now know: IF their is humanoidish life proper their: it'd look mostly like a bunch tightly compressed seed pods, and would avoid the redspot like the plague or would die, just about instantly, and no way in hell would it breath oxygen. methane maybe, but basically every gas that deletes human entrails is in it's atmosphere.
@futurebird a computer can never be a brain just as a painting can never be a landscape. they are fundamentally different, and that's that for all I'm concerned. could a computer model a brain to such an extent that it makes similar decisions? probably yes. would that make it a copy? no. can a computer ever have consciousness? idk but the more I have seen computers advancing in my lifetime the less I believe it will ever be possible.
@futurebird
Brains aren't some standardised platform like a CPU with a well-defined ISA. They're vines woven up against themselves, each in its own unique way.
And to date our tools for inspecting them are like studying a nuclear reactor from outside with infrared cameras, or cutting it into 2-D slices.
@futurebird Human brains don't work even remotely like a computer. The more we figure out about neurochemistry and bio-mechanics, the more obvious it is that we really don't understand human consciousness well enough to model it. That's why we've only begun to have success (if you can call it that) in modern AI when we gave up on human cognition and went to LLMs and big data-modes a computer is good at. I think the emulator would have to be unlike anything we've built so far, but not impossible.
Next 10 years, no, though I suspect we'll have something that can parrot responses for a few moments at a time shortly (discounting the terrible ones right now). In about 15, what Reynolds postulated as 'beta level' AI which are essentially stimulus response fakers modeled on real people.
True full brain upload? Eventually, perhaps- but it will take some dramatic changes and diverge rapidly (so much of our thoughts are based on recall speed).
@futurebird I would say it has to be POSSIBLE in principle, but that doesn't mean it has much chance of happening in practice - running new minds that are digital-native is just so much more efficient than emulation would be.
Add comment