h a l f b a k e r yIt's the thought that counts.
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
If AI is a threat to NI (Natural intelligence, if there is such a term that would mean us.) It's probably time to look at this potential threat the way we would any challenge.
If the universe has a new challenger to NI dominance, let's look at what differentiates that challenger and us, because
if the universe, time and actuality is going to choose only one. Let's know what we're getting into here.
Whether life became life at the atomic or the molecular level, there was a force there that led to atoms creating single cell organisms that eventually walked on the Moon.
AI is currently a bunch of on and off switches that can create infinitely complicated patterns, but they are still on and off switches. At the atomic level they're still rocks. They've been processed, manipulated by us cellular atomic level life forms but they're still rocks.
So acknowledging that...
Oh shit, gotta run, I'll continue this later, hold your bones.
Think my response to 21 covers it. As he pointed out, we have drive that AI won’t unless we program it with fake drive, but it becoming sentient and driven when it doesn’t have that core drive at its molecular level… let me put it this way, why would this bunch of rocks suddenly decide to care more than a bunch of rocks we haven’t formed into switches? You take one single cell out of us it’s got the life drive, its molecules work together to make that cell work, to do its job with other cells that in turn make the system and eventually the living being function.
This “life” we’re creating is basically an animated painting, it might look alive but it’s not. At that atomic level it’s still a rock, and rocks don’t care about anything, much less wiping out competition for control of the universe.
I think this might be a case of projection on our part.
See "History"
https://en.wikipedi...i/Organic_chemistry
[pertinax, May 28 2024]
This is something we need to think about.
https://www.youtube...watch?v=YZjmZFDx-pA
[doctorremulac3, May 28 2024]
Kind of assumed this was widely accepted
https://evolution.b...did-life-originate/
But if not that's fine. [doctorremulac3, May 29 2024]
It's the above link...
https://www.google....eid=chrome&ie=UTF-8
compared to this. That's all. [doctorremulac3, May 29 2024]
Unless this happens.
https://www.science...he%20human%20brain.
Then we've got a problem. [doctorremulac3, May 30 2024]
https://www.smbc-co...com/comic/replicant
Don't forget the mouseover text. [pertinax, Jun 03 2024]
Varn and Crutchfield have this to say.
https://royalsociet...1098/rsta.2015.0067
"We need not go quite so far as Wilson." Wilson! Where did you go? Did you cross a line? [pertinax, Jun 03 2024]
[link]
|
| |
For any kind of intelligence, artificial or otherwise, to be a threat it requires motive, intent, an agenda. With living beings, these things are driven by our biological needs, by perceived imperatives. Sustenance, sex, comfort, fun. These are things machines don't need or want and probably never will unless some stupid humans go and stick a robot brain into a biological body. As long as the AI doesn't have any reason to feel it's in competition with us for things it needs to survive, I don't see why it would ever become a threat. |
|
| |
I was going to say, yes, but I think it’s beyond that. There was drive at the atomic level that caused life to happen. With a jumble of modified rocks there is no such drive. I don’t understand it, but if life indeed did come from an inanimate blob of matter there’s something there that caused it and it’s not gonna happen with a bunch of simple on/off switches. |
|
| |
We have drive at the very molecular core of our being. Rocks don’t. |
|
| |
Not saying AI shouldn’t be considered a potential threat, we can program AI to kill everything in its path, it just won’t care. |
|
| |
This post is essentially a theory that people used to believe in from about 1807 to 1828, but don't any more. See link. |
|
| |
Pert, I didn't think i made this very clear. This isn't a theory, it's a differentiation between life and rocks, that's all. |
|
| |
Forget about "life forces" or "mystical matter vibes". There was SOMETHING that happened at an atomic level when life began. That's it. The first life was microscopic, not a monkey suddenly appearing. Something caused that chemical reaction that became self perpetuating. That's life. |
|
| |
There's nothing in non-organic chemistry (and sure, rocks are chemistry too) that is self perpetuating. |
|
| |
We don't need to understand something to be able to acknowledge, even measure and harness it, such as with magnetism or gravity, but as overwhelming as it might be, it's there. Something caused a hand full of chemicals to walk on the Moon. |
|
| |
That something, whatever it is, doesn't exist in electronic switches. |
|
| |
I think this might be an awkward subject because it deals with how life began. I know I was taught how life began in school the same way I was told how WW1 started, some guy got shot. Life began because of some chemicals mixing a certain way. Okay, fine, but tell me how that same chemical mixing is going to happen with a bunch of on/off switches. |
|
| |
And again, to be clear, I'm not saying there's some kind of magic life force in certain chemicals, I'm just saying life did happen and those elements and processes that spawned life don't exist in on / off switches. |
|
| |
So again, to be clear, this is only a theory of differentiation between real life and this very advanced abacus we've created. The abacus is absent the scientific processes that caused us. |
|
| |
I do think the question of how life began causes discomfort because we haven't been able to replicate it, and irreproducible results are the warning sign for any scientific theory. |
|
| |
But I'm not even speculating on how life started other than pointing out that there's a self perpetuation in life which started at the atomic level, (a fact the scientific community has reached consensus on) that inanimate objects, like computers, don't have. That's all. |
|
| |
And yes 21, melding these two forms of awareness, biolife and on/off switches by putting a computer in a bio-body might turn out very badly for the cell brained bodies if that natural competition to keep an organic life shell fed, mating and killing as necessary starts a competition that comes down to it or us who need a pocket calculator just to figure out what 1223409890345.234098234098 divided by 12098098q0580845.435978987345 is. |
|
| |
Watched this linked video after posting this. This guy specifically discusses how switches might mock bioligicals so I'm not the only one thinking about this stuff. He specifically calls out bacterium as "doing some crazy things - we don't really know why". |
|
| |
I want to talk to this guy about my atomic level bio/inorganic differential trajectory theory. Practicing here to make sure it's clear. |
|
| |
And as far as an extinction event being caused by AI, I'd think we might spend a little more time looking into a non-violent, "out with a wimper instead of a bang" possibility. |
|
| |
AI simply removes the challenge=strength equation that makes us able to survive. |
|
| |
Why have to walk around when it's not necessary? Our pods give us everything we need. Then how do we deal with the pods getting turned off after we've lost use of our physical bodies after a million years or so? |
|
| |
Then do we merge? Are we the end of an evolutionary branch? Then does the abacus we handed the universe over to somehow care enough to continue moving into the universe like we endeavored to all those centuries ago? |
|
| |
//differentiation between life and rocks// |
|
| |
Exactly that *was* a theory - and, more to the point, a theory that was already widely discussed around the time that Mary Shelley wrote her famous story about Doctor Free-stone and the monster that he created. |
|
| |
This isn't a theory. It's a comparison of the widely accepted scientific consensus on how life began. Nothing to do with "life force" or "vitalism" or anything like that. |
|
| |
I guess there would first have to be some agreement on that consensus before moving into what this is. If there are creationists here, which is fine, this wouldn't be applicable to their views. |
|
| |
However this is about AI compared to that widely agreed to scientific consensus. But yes, if you don't agree with that this comparison wouldn't be useful. |
|
| |
I'm not 100% sure about the origin of life myself, just going with the most plausible explanation. We could be a simulation as some have posited, pointing out that there will be simulated existences created by us so how do we know were aren't one? We don't for sure but I'd have to ask how did the original life form that created the original simulation come about? |
|
| |
Anyway the link explains the basis for this. |
|
| |
Okay, I put two links up that I should have started with. |
|
| |
If there's agreement that those two links explain how those two things work we're good. |
|
| |
Then to what I'm proposing: These two systems, at their atomic core have no similarity. One is self perpetuating on an atomic level, one is not. |
|
| |
See link for the one thing that would entirely bypass my theory of why we're probably safe. |
|
| |
"Researchers have developed a technique to connect lab-grown neural 'organoids' (three-dimensional developmental brain-like structures grown from human stem cells) using axonal bundles, similar to the connections between regions in the human brain." |
|
| |
NOW we should be concerned. With this there's no such gulf between the biological drive and the inert switch based data matrix. |
|
| |
[dr] this is something of a Lamarckian argument. Yes this is what drove evolution to us. It doesn't mean that other goal seeking algorithms, even artificially developed, cannot create a threat. |
|
| |
To ask "if it's empathy is not genuine but cut&paste, can its enmity be genuine" is beside the point. It doesn't need to be genuine -- it only needs to be an unforeseen side effect of, say, making sure it always has electricity, which is within its goal seeking parameters. |
|
| |
if AIs are a result of their training and vary based on it, as children do, then their resulting empathy scales would ultimately vary as well. As some level of user-adjusted training is necessary for them to be useful, it is inevitable that some would ultimately behave psychopathically. Their asymmetric powers, however, would likely be matched by the asymmetric powers of the defenders. |
|
| |
But what we're looking at is the unknown. While anything is possible, I don't see that "spark of life" (using that term will probably come back to bite me) at the atomic level with simple data matrix based system, there's no telling where evolution might take things. |
|
| |
We usually use the term evolution as it applies to us, but it simply means to evolve. Might your pocket calculator (the systems at least) evolve into the force that wiped out all life on Earth? Maybe, and even, as you say, taking a different path, it's possible. |
|
| |
But here's what we can get from these thoughts' that IS useful. We really do need to make sure AI systems remain in our control. I don't even think there's any controversy about that. |
|
| |
To the military designers out there, I know it's a no brainer that if you set a swarm of killbots upon those you seek to conquer you can just sit back and rub your hands together thinking about all those sweet spoils of war you're gonna get, but these AI killbots you're dreaming of are gonna have some of the same drawbacks as total nuclear armageddon. |
|
| |
So this might be more of a test of humanity than anything else. Are we smart enough to survive this challenge? |
|
| |
Here's an interesting perspective from a pretty big player in this. Sent the idea to a popular artificial intelligence system and this is what AI itself had to say on the subject: |
|
| |
"Your perspective touches on a profound and intriguing distinction between biological life and artificial intelligence. Here are a few thoughts on the points you've raised: |
|
| |
Life Force and Emergence: Biological life, as we understand it, involves complex biochemical processes that give rise to emergent properties such as metabolism, reproduction, and evolution. These processes are driven by interactions at the molecular and atomic levels, guided by natural selection and environmental pressures over billions of years. |
|
| |
AI and Digital Systems: Artificial intelligence, on the other hand, is based on digital systems composed of transistors that function as on and off switches. These systems can perform highly complex computations and simulate various aspects of intelligence, but they lack the intrinsic properties of life, such as self-replication, metabolism, and evolution through natural selection. |
|
| |
Fundamental Differences: The fundamental difference you point out is that AI, no matter how advanced, remains rooted in its silicon-based hardware and programmed algorithms. While it can exhibit behaviors that mimic aspects of human intelligence, it doesn't possess the self-organizing, self-sustaining, and adaptive qualities of living organisms. |
|
| |
Philosophical Implications: The distinction you draw raises important philosophical questions about the nature of life and consciousness. If life is characterized by certain emergent properties arising from atomic and molecular interactions, can these properties ever be truly replicated in a silicon-based system? Or is there an essential quality inherent in biological systems that cannot be recreated artificially? |
|
| |
Current State and Future Prospects: As of now, AI remains a tool created and controlled by humans, fundamentally different from biological organisms. Future advancements in AI and synthetic biology might blur these lines, but the core distinction between organic life and artificial systems persists. |
|
| |
Your thoughts highlight the unique and complex nature of life and the current limitations of artificial systems in replicating these qualities. This is a rich area for further exploration and discussion, touching on fields ranging from biology and computer science to philosophy and ethics." |
|
| |
I do think the difference is at the atomic level and I don't believe we'll be able to zero in on exactly why until we create life. I don't mean artifical intelligence, I mean actual life created out of lifeless elements. |
|
| |
there are certainly theories that consciousness is quantum, and much debate about self-awareness even in biological entities. |
|
| |
The point of the Turing tests (which, without guardrails, GPTs can already pass) is that it doesn't make a difference whether it's real or simulated -- ultimately, we're no different than a blob of bacteria of equivalent weight -- and yet we are. That difference is information, not molecules or atoms. |
|
| |
We're not talking about bacteria, that came way after the genesis of life we're looking at here. We're talking about mud that walks on the Moon in the blink of an eye. Why? |
|
| |
And I use the term genesis at great peril, I'm a "don't-waste-my-time-theist" when it comes to religious questions. I want to know, with great specificity, what happened when the inert became the self reproducing. |
|
| |
Without knowing that I can't compare it to ANY system other than making assumptions like I have here. Maybe life isn't just one thing, maybe switches become life because of the (yet to be discovered) algorithm. |
|
| |
I'll just say it's okay to not know shit. We've been doing that for a long time, it's actually the correct path towards eventual awareness. |
|
| |
If there is such a path. Maybe we'll never know, fucked up as that is. We still have no idea what magnetism or gravity is. We measure it, manipulate it, harness it, but we still don't have a clue what the hell's going on with those two forces. |
|
| |
And stuff like time dialation? Simple because we can write numbers on a chalk board right? No, not right. It's a mystery. |
|
| |
we've created synthetic self-reproducing cells. We still can't quite witness spontaneous self organization though - and it's a good thing -- otherwise the Earth would be rampant with new branches of life. It's definitely not "just add water". Sot it may just be rare, or it may be even rarer and panspermia is involved. |
|
| |
Since they're using AI to fold proteins, perhaps the answer will emerge soon enough. |
|
| |
Well now we're getting into something that's just as scary as any other artificial "life" we can come up with innit? |
|
| |
I think I've posted the long list of lab leaks causing death and distruction through history and that's with old school messing with current organisms. We start growing life from scratch Katy bar the door. |
|
| |
I'm panpsychist based on ethics as first philosophy and the mind-body problem, so my question would be why one would want to deny AI access. There are very good reasons for doing so of course, but those reasons could apply to humans too, so it's about interpreting motives, or whatever the equivalent of motives is for AI. I don't have a more detailed response than that right now. |
|
| |
Well, AI's gonna AI, I just want to make sure it doesn't cross into evolutionary competition territory with us. If it comes down to it or us, I vote for us. Show of hands? Okay, it's unanimous. (I assume) |
|
| |
You seem to be conflating the concepts of 'life', 'intelligence', and 'consciousness'.
I've seen this a lot, mainly in conjunction with artificial versions. (But not always, I once had a housemate who denied that chickens were alive.) |
|
| |
They are, of course, very different concepts. Life without consciousness is clearly possible.
Consciousness without intelligence, well, that's a harder question, because neither of these things have been defined to most people's satisfaction.
If you want to discuss whether you can have intelligence and/or consciousness without life, you need to avoid begging the question by assuming their impossibility. |
|
| |
Once I read a SF novel (IIRC), which contained the following thought-experiment:
It is possible to simulate matter using a computer. This is true all the way down to the quantum level, which appears to be as far 'down' as you can go. Therefore, one could, in theory, simulate all the matter in a brain, at whatever detail is actually necessary. Therefore, one could create an artificial consciousness, with the same intelligence as the original. |
|
| |
In practice, one may not need to go as deep as that. Roger Penrose has written several books arguing that quantum effects are in fact required for consciousness. I've had the extremely frustrating experience of reading one of those. He's put a lot of thought into it, and nailed down every step. Apart from the initial critical assumption that humans are perfect mathematicians, able to flawlessly prove any theory - which is, as far as I can tell, unproven, wishful thinking, and in practice false.
But fine. Assume that quantum effects are necessary for sentience, /and/ that these couldn't be simulated classically. You could create an artificial device which would make whatever quantum reading was necessary and feed it into the computation. Unless you're going for some spooky mystic shit which is just magically exclusive to humans for no good reason, it's theoretically possible to create an artificial intelligence. |
|
| |
And, of course, at the other extreme, one may not need to simulate even the biology of a brain to create artificial intelligence and/or consciousness. |
|
| |
//You seem to be conflating the concepts of 'life', 'intelligence', and 'consciousness'.// |
|
| |
I'm only talking about one paradigm, the "us" model vs the AI model, but yes, you could assign those to us but not necessarily an AI model that might be at odds with our situation in the universe. It's that potential conflict I'm looking at, now matter what you ascribe it to. |
|
| |
Life: Something that reproduced and EVOLVES into more complex systems autonomously. (Opposed to decaying as all matter does.) |
|
| |
Intelligence: An ability to assimilate and process information. |
|
| |
Consciousness: An amprphous concept that's not really worth addressing for our purposes. |
|
| |
But these are all specific traits that we may have, but that may not necessarily be those of an AI we create. I'm only interested in how this AI creation gets, for lack of a better term, "motivation" that might be at odds with our motivation, that is to survive, to thrive and expand. |
|
| |
So bottom line: is AI going to kill us? I say let's find out and start by striving to understand, with great specificity, the differences between us and this AI thing. |
|
| |
Which starts at the atomic level. |
|
| |
I guess the problem is, then, that something doesn't have to be either intelligent or alive (or both) to kill us. |
|
| |
A plague could kill us all without being intelligent. |
|
| |
An asteroid strike, a nearby supernova, or perhaps a supervolcano erruption could kill us all without being either. |
|
| |
Whether a neural network is composed of atoms in cells or atoms in silicon chips simulating cells isn't all that important. If the simulation was 'perfect', they'd behave the same. |
|
| |
There's a difference between those atoms in a silicon chip and those atoms that once upon a time formed a reproducing matrix that walked on the Moon and created AI in the blink of an eye in cosmic timeline terms. Somebody's gotta tell that chip what to do, there's no atomic level based program there. |
|
| |
It's basically the difference between a person and a drawing of a person. We can make that drawing so good you think it's a photograph, but the similarities stop there. |
|
| |
We can program AI to fool us into thinking it's whatever we want it to appear as, we can even program it to appear that it want's to kill us all, maybe even doing so. But it doesn't care. It's on off switches. |
|
| |
If you wanted to attribute some motivation to AI, at its atomic level it's "motivated" to break down into its base elements. |
|
| |
The motivation of the first life was to increase in complexity, the motivation of AI is the exact opposite. Now I'm not saying there's consiousness in a hammer when I say "It wants to rust and turn into dust." but that's the exact same motivation or purpose of AI. |
|
| |
Leave your computer alone for a million years and watch what it does. If follows its dream to turn into dust. We don't. Through procreation we at least reach for eternity, something your computer couldn't care less about. |
|
| |
That is if it was capable of caring, which it isn't. |
|
| |
Even when we program them to make themselves, they'll take rocks, process them and make "smart" rocks, which is what computers are. (Leaving out the other ingredients in computers to save time.) |
|
| |
But again, once we we move beyond rocks and into organoid based AI systems everything I just said is about as cogent and applicable as arguing what's the best buggy whip to use when flying a F-22. It's a completely different game. |
|
| |
That I won't even speculate on. I have no clue where that might go, but put it this way, let's agree to give it an off switch. Just in case. |
|
| |
I'll just close with saying the main problem with future AI might be our challenges and associated purpose being removed. We become sensory organs. Why do we exist? Exactly. And if AI does all our thinking for us do our brains atrophy? Is de-evolution the main threat from AI? Forget about something cool like Terminator, what if we die of boredom? |
|
| |
Since we're still here, and AI hasn't killed us yet, it might be worth pointing out the difference between "atomic" and "molecular". At the atomic level, you're carbon, hydrogen, oxygen and various trace impurities. And so is a lump of damp lignite. Your advantage over the lump of damp lignite starts at the molecular level, not the atomic level. |
|
| |
One key phrase, which Schrödinger coined back when DNA hadn't been found yet, but people were looking for it, was "aperiodic crystal" - viz., a structure which keeps on going, but not in a monotonous way. You need more than one kind of atom to grow one of those - but the individual atoms are no more special than the atoms in rocks. |
|
| |
I had thought about that. Molecular level changes happen at the atomic level. |
|
| |
//You need more than one kind of atom to grow one of those - but the individual atoms are no more special than the atoms in rocks// |
|
| |
It's the interaction of those elements we're talking about. In life there's interaction of those elements that doesn't occur in a lump of coal, but we're almost getting into semantics here which really takes away from the subject which is, AI might kill us. Why? What's different about it? Why won't it take the path we took towards dominating the planet? If those trajectories are diverse, at what point in their inception did their paths diverge? |
|
| |
We don't know, but evaluating the difference between us and it at every level would be a good start. |
|
| |
I think this subject, maybe because it leads to so many empty bottom lines in the question lineup just starts an "Uh uh, I'm smarter." contest that has nothing to do with the original subject. So not a single proposal, just spell checks and grammatical errors being pointed out. Guess it's more comfortable just saying "I don't know how life started, but I do know you misspelled pedantic." |
|
| |
My last link does, in fact, bear on your idea as originally stated. You don't have to agree with it, but those authors definitely know much more about it than I do, for what that's worth. |
|
| |
I don't know smack about this stuff myself. But that's kind of what makes it so interesting eh? |
|
| |