Comic Discussion > QUESTIONABLE CONTENT
WCDT strips 3671 to 3675 (5th to 9th February 2018)
SpanielBear:
--- Quote from: Aenno on 08 Feb 2018, 10:58 ---
--- Quote from: Stoutfellow on 08 Feb 2018, 06:20 ---
--- Quote from: Neko_Ali on 08 Feb 2018, 06:12 ---Mass Effect 2 & 3, EDI and Joker. EDI is the artificial intelligence that runs their ship and Joker is the pilot. At first she's always angry at him because he's wreckless and disorderly[.]
--- End quote ---
I would think wrecklessness would be a good thing in a pilot. :-)
--- End quote ---
Yes, with any other ship; but here we're talking about HER body, you know!
--- End quote ---
To clarify, I believe this is a pun based on a spelling mistake. The original term should have been (and has been corrected to) 'reckless', which does mean foolhardy and careless. The spelling mistake, 'wreckless', changed the meaning so the sentence became "Joker has never wrecked a space-craft" (He is wreck-less). Not crashing would be a positive attribute for a pilot to have...
Aenno:
Oh. That's as well is reading mistake from my part.
I'm sorry. Just forgot write spelling for "recklessness", so read it as it was supposed to be. :)
Case:
--- Quote from: Aenno on 07 Feb 2018, 14:43 ---
Human consciousness is, maybe, slow - it's basically an interface builded on hundreds of bugged and messy hormonal and instinctive systems, designed to long-forgotten stimules as well as current ones, and designed by a very bad designer (Mother Nature) with basic philosophy "this code worked adequatly when I submitted it, and I don't mind it's so buggy, we saving it for compatibility and support reasons"; so human consciousness just have no direct control to initiating body states.
But I can't see how "running action past a review function" is a bad thing.
--- End quote ---
Not entirely sure whether I get you correctly (or if I would agree if I did): I don't think consciousness is merely a 'review' system (Yes, it does have aspects of a review system, but ... that feels incomplete).
Consciousness 'feels' to me rather like something that initially (way back when, maybe even in one our non-human ancestors) started out as a biological equivalent to what programmers would call an "exception handling" routine to an already existing, very fast & adaptive primary cognitive system - an emergency system that can temporarily take over when the primary system encounters a situation it cannot solve (or a situation where it is prone to coming up with bad solutions) - and from those origins, it evolved and became better until ... one day, it got the idea that it was actually running the show.
Bit like Star Trek Voyager where the Emergency Medical Hologram first becomes the regular ship doctor because the regular ship doctor is dead, and in some situation even has to stand in for the captain (or the whole crew, was it?).
I'm interested in "Thinking, fast and slow" (but admit I haven't read it yet), because that's kind of how it feels to me - Science is the archetypical profession that demands "Slow down your thinking! Make your thoughts conscious!" (Srsly: At the root of it, that's even more fundamental than the Feynman-dictum (*), and that's already the "Scientific Method, in a nutshell"). But much of what I do as a physicist is not done by the slow, conscious, deliberative part of my mind - it's done by the quicksilvery "Dog chasing a stick"-part of my mind that can 'just see' a way to a solution, or completely forget time, space - even myself - during a calculation. And I am by no means convinced that I (as in 'the conscious part of myself') am the smarter of the pair, quite the opposite, in fact - it's more like ... it's way faster and smarter than I am, it has ancient wisdom, but in a funny way, it doesn't seem to be able to know what it's bad at, or when to stop. And when it looks in a mirror, it doesn't recognize itself.
I see 'my' job (again 'my' as in 'the conscious part of myself') as 'coming up with good questions for the part that is good at coming up with answers', as 'giving the dog a good scent and then getting out of its way'. When it comes back, my part starts again - looking at what the dog dragged up (patting its head, Good! Dog!)
And I think that another of the original functions of the conscious fraction of our mind was simulating another (proto-)human being. Suppose that for some reason - facilitating group interaction, predicting reactions of other group members, to stave of loneliness and despair, whatever - our brains found themselves in need of being able to speak to another, simulated human being, a mirror of sorts, in some situations. The 'exception handler' applied for the job and got it - and ... got good at it.
(And one day, it got this weird idea that it was actually running the show and all the rest was merely ... the bywork.)
Only then, all the other brains had their own 'simulated internal humans', too - probably the point we started inventing art in order not to go crazy. :-D
(*) "You must not fool yourself - and you are the easiest to fool"
Case:
Regarding the whole 'arousal/embarrassment/shame'-debate:
--- Quote from: Aenno on 07 Feb 2018, 15:09 ---<whole 'arousal/embarrassment/shame'-debate>
--- End quote ---
Some minor quibbles:
1) Human physical signs of arousal are not always coupled to an actual experience of arousal - Every guy past puberty knows that (Well, unless I'm the only dude on Earth who wakes with morning wood, yet breathing quite normally and thinking more of going for a wee rather than, say, a raging urge to either masturbate or search the premises for any potential mating partners that may happen to be around and/or have snuck into his flat overnight). Men get raging erections in the most impossible of situations, feeling exactly zero sexual arousal - the 'combat boner' is an example (That's actually part of how men absolutely can get raped - same as how a woman doesn't enjoy and/or consent to intercourse whenever her body decides it's time to get wet). And that's talking about what is probably the simpler part of the species.
2) IIRC, Human children start displaying signs of shame long before they first experience sexual arousal - Kids ban their parents (and especially the 'desire target'-parent) from the bathroom long before puberty.
3) We've heard that AI's can experience a sort of VR-sex amongst each other. If that is sex in any definition we would recognize, it involves at least the simulated experience of sensual stimuli and involuntary reactions, regardless of whether their everyday 'chassis' are capable of experiencing those sensations.
4) RetCon: Bubbles chassis is a recent experimental model. Maybe someone at Skunkworks got tired of designing horny little psychopaths with built-in lasers who are way too much into niche human fetishes for their own good, and decided that "if the bugger are so keen on it, they should start making their own kink"?
5) RetCon: Humans are smart apes, but apes we are. No matter what we do, our first basic questions are always "Can it eats me?" "Can I eats it?" "Can I mate with it?". There's something funny - and comforting - in knowing that at least 5-15% of our most recent paradigm-shattering invention The! Internet! is dedicated exclusively to pron. Even someone as otherworldly as AI-inventor Chatham would have expected that at some point, AI would be put into human-form chassis, and from there, it's a giant leap for AI-kind but a quick stroke for ... (trying to come up with a good way to end that one).
So it would make sense that the people who wrote the first versions of AI "sensual faculty software .dll's" would add at least some placeholder entries for 'sexual pleasure' (or even approximations), even if the bodies capable of feeling those sensations didn't yet exist - simply because knowing their fellow humans, they knew those bodies would exist at some point.
TL;DR - It's not impossible for Bubbles to feel shame without feeling arousal. It's not impossible for Bubbles to feel aroused even if her body would (yet) lack all the sensory faculties for lovemaking.
P.S.: Are you familiar with Neal Asher's Polity Universe? I think you might like them. There's a type of "AI" called Golem that kind of fits some of your ideas - an android chassis with a non-organic cognitive core that can be host to both 'natural' AI's as well as recorded human minds (Asher subtly hint that over time, the difference may start to become less and less important). The trick with the Golem's is that the chassis run an emulation of human basic emotion and also mimick other human peculiarities - pretending that their joints have the same range as ours when they don't, gait, sweat, body odor, etc.
He keeps coming back to the question of "how real simulated emotion is" over several books, culminating in one part where a formerly human side-character is re-awakened after death into an AI chassis, with the ability to switch off his human emotions - and starts wondering not how human he is, but how real his identity is. Is he still the real Gant that he feels he is, when he knows he is so different, so much more in some respects.
Aenno:
--- Quote ---Not entirely sure whether I get you correctly (or if I would agree if I did): I don't think consciousness is merely a 'review' system (Yes, it does have aspects of a review system, but ... that feels incomplete).
Consciousness 'feels' to me rather like something that initially (way back when, maybe even in one our non-human ancestors) started out as a biological equivalent to what programmers would call an "exception handling" routine to an already existing, very fast & adaptive primary cognitive system - an emergency system that can temporarily take over when the primary system encounters a situation it cannot solve (or a situation where it is prone to coming up with bad solutions) - and from those origins, it evolved and became better until ... one day, it got the idea that it was actually running the show.
--- End quote ---
AFAIK it's true. Consciousness is, in a nutshell (very, very small nutshell) a system allowing to break a program (instinct) and find a way to solve a problem. It's not faster or slower that any other part by mechanism, but it's far more distracted, and have very little resource in its disposal. You can make your consciousness work fast - you'll just need to concentrate it on "this particular problem" to solve. And it's actually very difficult skill, and very tiresome one. And, as we haven't full control, we can't actually drag out every problem into consciousness.
Mental trauma is interesting example. It actually "heal itself" slowly (as nothing really happens "itself" this meaning it healing itself subconsciously). Consciousness approach, if you have correct skills and able to defeat the suffering in process, is FAR faster.
Or take learning process. As people gasp learning skills and required concentration, learning became far faster with consciousness approach.
Thing with subconscious is that it uses shortcuts. Consciousness can use it as well, as long as you have skills. I mean, it's like... "Subconscious is very fast in math. You just need to look into nightsky, and it's already done - there are a lot of stars here!"
--- Quote from: Case on 08 Feb 2018, 12:26 ---Regarding the whole 'arousal/embarrassment/shame'-debate:
--- Quote from: Aenno on 07 Feb 2018, 15:09 ---
--- Quote from: SpanielBear on 07 Feb 2018, 15:02 ---But to answer you question "How do AI's have uncontrolled emotional responses" (I think I've got that right?) with "Because Magic" would be completely unsatisfying.
--- End quote ---
Sorry, no, you haven't got that exactly right.
Question is "How do AIs have uncontrolled emotional responses by every detail resembling human sexual arousment."
Because if there is a emotion that have less social component that direct sexual arousment, I don't know it. It can be emulated, sure - but WHO emulate it in Bubbles?
--- End quote ---
Some minor quibbles:
1) Human physical signs of arousal are not always coupled to an actual experience of arousal - Every guy past puberty knows that (Well, unless I'm the only dude who wakes with morning wood and doesn't have to suppress the raging urge to either masturbate or have sex with an available and willing partner that happens to be around). Men get raging erections in the most impossible of situations, feeling exactly zero sexual arousal - the 'combat boner' is an example. And that's talking about what is probably the simpler part of the species.
2) IIRC, Human children start displaying signs of shame long before they first experience sexual arousal - Kids ban their parents (and especially the 'desire target'-parent) from the bathroom long before puberty.
3) We've heard that AI's can experience a sort of VR-sex amongst each other. If that is sex in any definition we would recognize, it involves at least the simulated experience of sensual stimuli and involuntary reactions, regardless of whether their bodies are capable of experiencing those sensations.
4) RetCon: Bubbles chassis is a recent experimental model. Maybe someone at Skunkworks got tired of designing horny little psychopaths with built-in lasers who are way too much into niche human fetishes for their own good, and decided that "if the bugger are so keen on it, they should start making their own kink"?
5) RetCon: Humans are smart apes, but apes we are. No matter what we do, our first basic questions are always "Can it eats me?" "Can I eats it?" "Can I mate with it?". There's something funny - and comforting - in knowing that at least 5-15% of our most recent paradigm-shattering invention The! Internet! is dedicated exclusively to pron. Even someone as otherworldly as AI-inventor Chatham would have expected that at some point, AI would be put into human-form chassis, and from there, it's a giant leap for AI-kind but a quick stroke for ... (trying to come up with a good way to end that one).
So it would make sense that the people who wrote the first versions of AI "sensual faculty software .dll's" would add at least some placeholder entries for 'sexual pleasure' (or even approximations), even if the bodies capable of feeling those sensations didn't yet exist - simply because knowing their fellow humans, they knew those bodies would exist at some point.
TL;DR - It's not impossible for Bubbles to feel shame without feeling arousal. It's not impossible for Bubbles to feel aroused even if her body would (yet) lack all the sensory faculties for lovemaking.
--- End quote ---
Well... again, it's complicated.
1) About "most impossible situations to have erections". The very problem with human body is that we haven't any hormone with only one task. My favorite example is oxytocin. Look into wiki to a list of tasks!
How do erection work, with a great simplification? Humans have enzyme, horrible molecule called cyclic guanosine monophosphate. There are receptors in human blood system that have a simple program - "if cGMP hit you, wide blood vessel". And cGMP is synthesized normally, and degrading with another horrible molecule, phosphodiesterase. Phosphodiesterase V (PDE-5) is concentrated in pelvic cavity and supposed to destroy cGMP as fast as it needed. If sexual arousal happen, PDE-5 synthesis get low, pelvic cavity organs (including corpus cavernosum) had their blood vessels widen, blood pressure up, voila - erection.
Imagine by some incident you have cGMP synthesis rise or PDE-5 synthesis low. PDE-5 wouldn't be enough to destroy cGMP. That's how Viagra works, by the way, it inhibit PDE-5 synthesis. Voila - erection.
Problem is, cGMP also using in apoptosis (by-designed death for cells in the body), AND in translation light into electricity in your eyes, AND in fight-and-flight reaction. And body is stupid. It never can access situation. "cGMP level rise = erection".
Biochemistry is fun, but EXTREMELY messy. So yeah, not every erection is arousal. And not every oxytocin blast is arousal. And not every nipple erection is arousal.
2) Yes, because children are teached to do this things, it's not instinctive reaction.
3) We actually do know mechanism. It's high-speed package exchange.
4) Continuity error: It was directly declared that Pintsize program and Bubble program was running parallel, and Pintsize's one was closed for humanitarian reasons.
5) Definitely possible, and I believe was done. But then it's the same thing with robotic drunkenness. In a nutshell, as far it was shown, it's a reaction that AI summons on himself consciously. "I want to be drunk for social reasons, so I'm downgrading processor cycles and apps I have would initiate "drunk" behavior".
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version