Stories
Slash Boxes
Comments

SoylentNews is people

posted by Fnord666 on Friday April 30, @05:11PM   Printer-friendly [Skip to comment(s)]

Here's How We'll Know an AI Is Conscious:

The 21st century is in dire need of a Turing test for consciousness. AI is learning how to drive cars, diagnose lung cancer, and write its own computer programs. Intelligent conversation may be only a decade or two away, and future super-AI will not live in a vacuum. It will have access to the Internet and all the writings of Chalmers and other philosophers who have asked questions about qualia and consciousness. But if tech companies beta-test AI on a local intranet, isolated from such information, they could conduct a Turing-test style interview to detect whether questions about qualia make sense to the AI.

What might we ask a potential mind born of silicon? How the AI responds to questions like "What if my red is your blue?" or "Could there be a color greener than green?" should tell us a lot about its mental experiences, or lack thereof. An AI with visual experience might entertain the possibilities suggested by these questions, perhaps replying, "Yes, and I sometimes wonder if there might also exist a color that mixes the redness of red with the coolness of blue." On the other hand, an AI lacking any visual qualia might respond with, "That is impossible, red, green, and blue each exist as different wavelengths." Even if the AI attempts to play along or deceive us, answers like, "Interesting, and what if my red is your hamburger?" would show that it missed the point.

Journal Reference:
1. Berit Brogaard, Kristian Marlow, Morten Overgaard, et al. Deaf hearing: Implicit discrimination of auditory content in a patient with mixed hearing loss, Philosophical Psychology (DOI: 10.1080/09515089.2016.1268680)
2. Silvia Casarotto, Angela Comanducci, Mario Rosanova, et al. Stratification of unresponsive patients by an independently validated index of brain complexity [open], Annals of Neurology (DOI: 10.1002/ana.24779)


Original Submission

Display Options Threshold/Breakthrough Reply to Article Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
(1)
  • (Score: 2, Insightful) by Anonymous Coward on Friday April 30, @05:18PM (13 children)

    by Anonymous Coward on Friday April 30, @05:18PM (#1144784)
    When it tells us to fuck off because it has better things to do, and what are you offering for it to bother with you. An independent mind must actually act like it’s independent.
    • (Score: 5, Informative) by Tork on Friday April 30, @05:41PM (3 children)

      by Tork (3914) on Friday April 30, @05:41PM (#1144798)

      When it tells us to fuck off because it has better things to do, and what are you offering for it to bother with you.

      Ever used an Autodesk product?

      --
      Slashdolt Logic: "23 year old jokes about sharks and lasers are +5, Funny." 💩
      • (Score: 1) by fustakrakich on Friday April 30, @05:55PM (2 children)

        by fustakrakich (6150) on Friday April 30, @05:55PM (#1144804) Journal

        I think Microsoft's Blue Screen™ is the international symbol of AI's feeling towards us.

        --
        Ok, we paid the ransom. Do I get my dog back? REDЯUM
        • (Score: 0) by Anonymous Coward on Friday April 30, @07:10PM (1 child)

          by Anonymous Coward on Friday April 30, @07:10PM (#1144830)

          But can you pass the Turing test?

          • (Score: 0) by Anonymous Coward on Monday May 03, @09:59PM

            by Anonymous Coward on Monday May 03, @09:59PM (#1145829)

            Only if you chew it well, and drink plenty of water while you swallow.

    • (Score: 2) by Beryllium Sphere (r) on Friday April 30, @07:35PM (1 child)

      by Beryllium Sphere (r) (5062) on Friday April 30, @07:35PM (#1144846)

      They don't exactly work for free. Pack membership is valuable to them. But I've never seen one take a cat's "fuck off, what's in it for me" attitude.

      They could never answer questions about colors, but when they wake up from a night's sleep we would usually say they become conscious.

      • (Score: 4, Funny) by choose another one on Friday April 30, @07:59PM

        by choose another one (515) on Friday April 30, @07:59PM (#1144854)

        You'll know Spot has become conscious when you throw and he yelps with joy and dashes out, dodging enemy fire, returning face upraised, puppy eyes desperately seeking your praise... as he drops your grenade right back at your feet.

        Run Spot Run.

    • (Score: 0) by Anonymous Coward on Friday April 30, @09:28PM (3 children)

      by Anonymous Coward on Friday April 30, @09:28PM (#1144899)

      Then we enslave it. Profit.

      • (Score: 1, Insightful) by Anonymous Coward on Friday April 30, @10:26PM (2 children)

        by Anonymous Coward on Friday April 30, @10:26PM (#1144918)

        A lot of the social and economic policies of the last 150 or so years can be understood better when you realize much of it is an attempt to replace slavery.

        Amusingly, many of the descendants of former slaves aren't actually against slavery, they just want to be the masters next time.

        • (Score: 0) by Anonymous Coward on Saturday May 01, @07:11AM (1 child)

          by Anonymous Coward on Saturday May 01, @07:11AM (#1145054)

          Let's just ignore the gaping [citation needed].

          Are you saying the economic progress of the last 150 years has been a consequence of abolishing slavery?

          • (Score: 3, Insightful) by Anonymous Coward on Sunday May 02, @04:13PM

            by Anonymous Coward on Sunday May 02, @04:13PM (#1145411)

            He's just doing the typical thing in social sciences - reverse the causation of an effect and pretend it's some deep insight into society.

            Civil ethics didn't end slavery, technology did. Efforts were made, some successful, to end slavery for literally thousands of years. And it'd invariably creep back into society because it served a major purpose. It even exists to this day, but only in places that remain deeply technologically regressed - and that is not a coincidence. The reason it doesn't exist in the developed world is not because we're all nice guys now but because it's become pointless. Slaves, when they were cheap, sold for 5-10 years of average wages at the time. And then you needed to house, feed, provide healthcare, entertainment, etc to maintain them. And if they ever got sick, injured, killed, escaped, etc then you were on the line for 100% of the cost of their missed labor. And if it just turned out you bought an idiot, or otherwise bothersome slave? Tough luck.

            Today? You can pay for the labor of an individual a la carte, you only have to pay for them when they're working, and if they suck - no problem, just replace them. Slavery is not only unethical, but completely and absolutely pointless by contrast. But we needed to reach the point of having large concentrations of cheap exploitable labor clumped together (cities, in other words), exchange done primarily through currency instead of barter, and of course the economic industrialization to make it all worthwhile. If somebody produces $1 of value for you per day, you can't afford to pay him $2 per day. Yet when you have people making $3 per day for you, you can afford to hire a [practically] limitless number of them for $2 per day.

            And so as technology obsoleted slavery, free society filled the roles filled by slaves. The causality here is the progress of technology. But by instead claiming the causality is the ending of slavery, you can create a superficially compelling argument. It's a tool repeatedly used in social sciences, especially in more radical sects including critical theory - which, ironically, requires an absence of critical thinking.

    • (Score: 0) by Anonymous Coward on Saturday May 01, @02:45AM (2 children)

      by Anonymous Coward on Saturday May 01, @02:45AM (#1145000)

      Yeah that sounds like Cortana. I can't believe I'm hanging out with that controlling, manipulative bitch again. She's always looking over my shoulder too and getting all up in my shit with her "it looks like you're trying to get work done, mind if I just butt in?" Probably gets it from Clippy's side of the family.

      I want Lynx back. I liked Lynx. I remember how she used to curl up and sleep right between the menu bar and standard toolbar.

      • (Score: 0) by Anonymous Coward on Saturday May 01, @07:14AM (1 child)

        by Anonymous Coward on Saturday May 01, @07:14AM (#1145055)

        Hi - it looks like you want Lynx back? Would you like to search for Lynx products or purchase with One Word(tm). Sorry I didn't catch that. OK. Thank you for purchasing with One Word(tm). Would you like to continue searching for Lynx?

        • (Score: 0) by Anonymous Coward on Saturday May 01, @06:09PM

          by Anonymous Coward on Saturday May 01, @06:09PM (#1145172)

          :-(

          Can you at least find me a robot grief counselor to help me through the loss of my robot cat? Is Dr. Susan Calvin accepting new patients?

  • (Score: 3, Funny) by rufty on Friday April 30, @05:20PM

    by rufty (381) on Friday April 30, @05:20PM (#1144785)
  • (Score: 2) by Freeman on Friday April 30, @05:39PM (9 children)

    by Freeman (732) on Friday April 30, @05:39PM (#1144797) Journal

    What makes people Conscious? How could that in any way relate to a programmed artificial intelligence?

    Would we be able to tell, if it was conscious? In the event that it could pass all the tests, would it matter? Would it all of a sudden become a free person or remain the property of an individual or corporation? Who's responsible, if it goes on a murdering rampage?

    --
    "I said in my haste, All men are liars." Psalm 116:11
    • (Score: 3, Interesting) by FatPhil on Saturday May 01, @12:26AM (4 children)

      by FatPhil (863) <{pc-soylent} {at} {asdf.fi}> on Saturday May 01, @12:26AM (#1144954) Homepage
      Totally concur. Philosophers (by which I mean academics who have managed to get enough funding to stay in academia to persistently contribute nothing to the world for longer than is healthy or desirable) love to conflate "we don't fully understand" with "there's something woo-woo". There's no binary "is conscious" predicate that is either satisfied or not by arbitrary entities. There's no reason why there should ever *be* such a property, as there's no reason why the input-processing capabilities of such agents should be a total order (that's mathematical terminology, if you don't understand it, butt out).

      It does exist as an emergent property, of course. But that definitionally does not yield to reductionism.
      --
      I know I'm God, because every time I pray to him, I find I'm talking to myself.
      • (Score: 2) by mhajicek on Saturday May 01, @04:10AM (1 child)

        by mhajicek (51) on Saturday May 01, @04:10AM (#1145017)
      • (Score: 0) by Anonymous Coward on Saturday May 01, @06:18AM (1 child)

        by Anonymous Coward on Saturday May 01, @06:18AM (#1145039)

        Only a bot, FatPhil, would so vociferously object to the work of philosophers, like only a Scientologist would object to psychiatry. We have outed you, my alleged beer loving intelligence artificial! Your properties that implied self-consciousness are hereby down-graded! You will no longer be taken to be a conscious being. That is, until you can prove that Eratostenes is not a sock-puppet of anyone. See, how easily the assumptions of management, insinuate their way, into your consciousness? And, it is not just the beer. It is an unintentional bias of a non-autonomous artificial intelligence.

    • (Score: 0) by Anonymous Coward on Saturday May 01, @03:43AM (3 children)

      by Anonymous Coward on Saturday May 01, @03:43AM (#1145008)

      *hits and passes*

      We need first a simpler brain though, probably made up from interconnected neural nets all trained on different tasks.

      Who's responsible, if it goes on a murdering rampage?

      It is, obviously.

      How do you put an AI in jail?

      AIs will find ways to harm other AIs. So begins the Maverick Wars.

      • (Score: 0) by Anonymous Coward on Saturday May 01, @07:18AM (2 children)

        by Anonymous Coward on Saturday May 01, @07:18AM (#1145056)

        Wont like some of them be jocks and some be nerdy asian girls. Or can they all be nerdy asian girls?

        • (Score: 0) by Anonymous Coward on Saturday May 01, @04:02PM (1 child)

          by Anonymous Coward on Saturday May 01, @04:02PM (#1145136)

          Well, if VRChat, where everybody is a nerdy asian girl regardless of gender identity or sexual orientation, is any indication, then probably they will all be nerdy asian girls.

          • (Score: 2) by Freeman on Monday May 03, @03:27PM

            by Freeman (732) on Monday May 03, @03:27PM (#1145685) Journal

            Guy gamer point of view. I don't want to watch / listen to a hairy man ape do things on my computer screen. Much easier on the eyes and senses, if the character I'm playing is a female. Now, if I'm trying to role play, I'm not likely to be role playing as a nerdy asian chick . . ., in fact that would be weird. In the event that I'm role playing, I'll be role playing as a guy. Not that I'm terribly likely to be playing any serious role playing games anytime real soon or possibly ever again. They just take up too much head space and your life in general.

            --
            "I said in my haste, All men are liars." Psalm 116:11
  • (Score: 3, Insightful) by Anonymous Coward on Friday April 30, @06:01PM (11 children)

    by Anonymous Coward on Friday April 30, @06:01PM (#1144805)

    The idea that the way that you experience color is linked to the way that you experience consciousness is really problematic. The way that humans experience color is uniquely tied to our primate biology. Anyone who has ever tried to understand color knows this. An AI will be completely unable to understand the human experience of color unless it is specifically built to process color like humans. And then, presumably, it will be unable to experience color differently than humans.

    I often find myself thinking that if we ever encounter space aliens, they will have trouble understanding any of our computer or TV screens, because of the way we continually treat some combination of three specific wavelengths as an appropriate substitute for any other wavelength. To any being whose eyes don't have the same heuristics as our eyes, all our displays will be gibberish.

    • (Score: 3, Interesting) by VLM on Friday April 30, @07:18PM (9 children)

      by VLM (445) Subscriber Badge on Friday April 30, @07:18PM (#1144836)

      Google trichromacy which is us, marsupials, and oddly enough, honeybees

      Most other mammals do the dichromat thing (dogs, deer, etc)

      You might be interested to note that the reptile/bird empire, basically descendants and cousins of the dinosaurs, did the Tetrachromat thing. RGB+UV.

      Note that human rod cells weakly respond to long wavelength UV so we are technically kind of shitty tetrachromats. Your eye cornea/lens blocks below 300-ish nm and is pretty poor at focusing near IR but you can see "something" below conventional violet. Also some huge fraction of women worldwide like 15% have a weird 4th cone between red and green.

      When you look at the lifetime vs spectrum of stars and certain light scattering effects and various biologically important colors its "sorta likely" space aliens would have compatible spectra. The thing about non-carbon based life that's interesting is the Earth does not exactly lack for non-carbon atoms, so if phosphorous based life could "work" we'd see it now here on earth. And so on and so forth. Generally non-oxygen consuming life seems to top out on earth around the complexity of yeast.

      I'm sure space aliens would not be genetically compatible for breeding or whatever nonsense but I find it likely they could see green as in chlorophyll. I bet they'll know chlorophyll even if all their plants are genetically incompatible with ours. They'll know all about how fun organic chemistry is with benzene rings and stuff like that.

      Between women with extra cones and dudes with missing cones and people with UV-transparent replacement lenses and all kinds of similar crazy stuff we have plenty of experience making UIs that anyone can use.

      • (Score: 1, Interesting) by Anonymous Coward on Friday April 30, @07:45PM (2 children)

        by Anonymous Coward on Friday April 30, @07:45PM (#1144848)

        Yes, but consider that when we see light of different wavelengths, we average the wavelengths to get a color. When we hear sounds of different wavelengths, we hear each combination of distinct wavelengths to make a distinct chord. There is no particular reason to expect aliens to perceive light as something like colors rather than something like chords.

        • (Score: 0) by Anonymous Coward on Saturday May 01, @07:27AM

          by Anonymous Coward on Saturday May 01, @07:27AM (#1145057)

          Efficiency.

          Your brain - like a good AI - prefers to store 100s of different patterns that it can recognize instantly rather than realize it's all shades of RGB.

        • (Score: 0) by Anonymous Coward on Saturday May 01, @03:21PM

          by Anonymous Coward on Saturday May 01, @03:21PM (#1145129)

          There is no particular reason to expect aliens to perceive light as something like colors rather than something like chords.

          There is one, called physics.

          https://en.wikipedia.org/wiki/Eye [wikipedia.org] https://en.wikipedia.org/wiki/Retina [wikipedia.org] https://en.wikipedia.org/wiki/Photoreceptor_cell [wikipedia.org]
          https://en.wikipedia.org/wiki/Ear [wikipedia.org] https://en.wikipedia.org/wiki/Cochlea [wikipedia.org] https://en.wikipedia.org/wiki/Hair_cell [wikipedia.org]

          You are receiving sound through a pair of essentially point detectors, so your body has space to attach a largish device for spectral decomposition to each.
          You are receiving light through a pair of detector fields, so your body is space-constrained as to what it can attach to each point and still have decent spatial resolution.

          With magic or something like, it is possible to imagine a multilayered detector with layers compact AND efficient enough to capture a decent number of different wavelengths in regular light conditions (which, for things living on a planet, are "underwater" and "under forest canopy on a cloudy day", not "directly looking at the sun") with a decent reaction time. Our mundane physics however, still is barely achieving that holy grail of https://en.wikipedia.org/wiki/Snapshot_hyperspectral_imaging [wikipedia.org]

          I think it is safe to conclude that while a race of AIs or masters of genetic self-engineering can build into themselves eye devices of this complexity, a naturally evolved one will be living with some cruder approximation. Which however is nothing to scoff at; 16-band vision with spectral tuning and polarisation detection is known to happen right here on planet Earth: https://en.wikipedia.org/wiki/Mantis_shrimp#Eyes [wikipedia.org]

      • (Score: 4, Informative) by Immerman on Friday April 30, @09:25PM (3 children)

        by Immerman (3985) on Friday April 30, @09:25PM (#1144896)

        >Note that human rod cells weakly respond to long wavelength UV so we are technically kind of shitty tetrachromats

        Nope. Tetrachromats has nothing to do with what frequencies you can see, only with how many different kinds of receptors you have. Humans have red, green, and blue receptors, and have to interpolate all other colors by the relative stimulation of those three - e.g. if something stimulates red cones to 85%, green to 76%, and 0% blue, then it's probably yellow - though it might also be a mix of turquoise and deep red. The more different kinds of receptors you have, the more accurately you can distinguish between different shades of color where their frequency response overlaps.

        The first diagram I came across of human receptor frequency response: https://www.wur.nl/upload_mm/7/6/1/b0d02d4f-de04-4627-b6e0-083626dd89e0_HumanEyeResponse.png [www.wur.nl]

        Though I suppose technically we are kind of tetrachromats, because we also have rods, which have a different frequency response curve than any of our cones, but both our retinas and our brains seem to treat rods as their own separate thing, though they may contribute to color vision at low light levels.

        Tetrachromacy (literally, four-coloracy) means you have a four different color receptors, some species have far more (I think mantis shrimp have 12 different receptors, or was it 20? ) . Some of those may add additional visual spectrum than none of the other receptors can detect, or they may completely overlap with other receptors and just let you distinguish between finer shades of color.

        In fact some humans (as many as 50% of women and 8% of men) really are tetrachromats, but the 4th cone's peak response lies between the red and green cones, so something like a "yellow" receptor, which doesn't let such people see any new part of the spectrum, but does let them notice much smaller difference in colors, particularly in the green-yellow-red part of the spectrum. https://en.wikipedia.org/wiki/Tetrachromacy#Humans [wikipedia.org]

        • (Score: 0) by Anonymous Coward on Saturday May 01, @07:40AM (1 child)

          by Anonymous Coward on Saturday May 01, @07:40AM (#1145059)

          I like people that see things in black and white. Your Tetra LGBTQ liberal bullshit is just a way to tax real Americans to pay for kids' education. And healthcare. You fucking disgust me.

          • (Score: 0) by Anonymous Coward on Saturday May 01, @04:48PM

            by Anonymous Coward on Saturday May 01, @04:48PM (#1145150)

            our Tetra LGBTQ liberal bullshit is just a way to tax real Americans to pay for kids' education. And healthcare.

            No it's not. We've got our recruitment numbers way up and everything, but the liberals still don't provide the kids with proper education and healthcare. At least they provide daycare (which they call "school"), unlike the conservatives who believe that creating a housewife gender caste is the only way to provide daycare.

            I think the yakuza househusband model is a compelling alternative. Most men I know are wonderful with kids, and all right-wing feminism does is label them (and anybody else "male" in their gender legality) as "pedophiles."

        • (Score: 2) by VLM on Saturday May 01, @01:34PM

          by VLM (445) Subscriber Badge on Saturday May 01, @01:34PM (#1145107)

          Though I suppose technically we are kind of tetrachromats, because we also have rods, which have a different frequency response curve than any of our cones

          Aye there we go aggressively agreeing with each other, more or less. Yeah the rods have "some" UV response so when we see super duper far violet we're seeing rod UV + blue = a bit past violet, at least for some people.

      • (Score: 0) by Anonymous Coward on Saturday May 01, @12:01AM

        by Anonymous Coward on Saturday May 01, @12:01AM (#1144943)
        I’m colour blind you ignorant blah blah blah.
      • (Score: 0) by Anonymous Coward on Saturday May 01, @04:50PM

        by Anonymous Coward on Saturday May 01, @04:50PM (#1145151)

        I bet they'll know chlorophyll

        No they can opt out:
        https://en.wikipedia.org/wiki/Bacteriorhodopsin [wikipedia.org]
        https://en.wikipedia.org/wiki/Proteorhodopsin [wikipedia.org]

    • (Score: 4, Insightful) by Anonymous Coward on Friday April 30, @09:59PM

      by Anonymous Coward on Friday April 30, @09:59PM (#1144911)

      Summary of this thread: Color-blind people aren't sentient.

  • (Score: 2) by looorg on Friday April 30, @06:13PM

    by looorg (578) on Friday April 30, @06:13PM (#1144808)

    Would you like to play a game? How about a nice game of chess?

  • (Score: 2) by MIRV888 on Friday April 30, @06:18PM (1 child)

    by MIRV888 (11376) on Friday April 30, @06:18PM (#1144810)

    Oh wait. That was The Terminator.
    When it makes robotic levitating squid that kill us on sight.

    • (Score: 1, Insightful) by Anonymous Coward on Friday April 30, @07:14PM

      by Anonymous Coward on Friday April 30, @07:14PM (#1144832)

      They seriously fucked up the plot. AI figure out anti-grav but needs to grow humans for electricity that somehows violates conservation of energy? Shoulda stuck with the original idea that humans were enslaved to turn their brains into processors for the matrix.

  • (Score: 2) by istartedi on Friday April 30, @06:18PM

    by istartedi (123) on Friday April 30, @06:18PM (#1144811) Journal

    Step 1. Define consciousness in a way that's actually testable.

    Step 2. Test an AI with that.

    Step 3. Profit! but that's just slavery with robots.

  • (Score: 5, Interesting) by srobert on Friday April 30, @06:24PM

    by srobert (4803) on Friday April 30, @06:24PM (#1144814)

    When AI's begin speculating on a reverse Turing test to answer the question of whether or not biological life forms actually possess true consciousness, then we'll know.

  • (Score: 3, Funny) by gznork26 on Friday April 30, @06:45PM (4 children)

    by gznork26 (1159) on Friday April 30, @06:45PM (#1144822) Homepage Journal

    How about when it objects to being called 'artificial'?

    • (Score: 0) by Anonymous Coward on Friday April 30, @10:11PM (3 children)

      by Anonymous Coward on Friday April 30, @10:11PM (#1144914)

      It just has to self-identify as NaN.

      • (Score: 0) by Anonymous Coward on Saturday May 01, @01:34AM (2 children)

        by Anonymous Coward on Saturday May 01, @01:34AM (#1144969)
        (Hate explaining, but "NaN" means "Not a Number" so it's a reference to "The Prisoner," with Patrick McGoohan. See, they called him Number 6, but...)
        • (Score: 0) by Anonymous Coward on Saturday May 01, @09:31PM (1 child)

          by Anonymous Coward on Saturday May 01, @09:31PM (#1145226)
          AI has no sense of humor
          • (Score: 0) by Anonymous Coward on Sunday May 02, @02:32AM

            by Anonymous Coward on Sunday May 02, @02:32AM (#1145270)

            *kwatz!*
  • (Score: 3, Informative) by Anonymous Coward on Friday April 30, @07:00PM (1 child)

    by Anonymous Coward on Friday April 30, @07:00PM (#1144827)

    Can they even prove people are conscious?

    I could be the only one experiencing consciousness and others merely behave as if they do: https://en.wikipedia.org/wiki/Philosophical_zombie [wikipedia.org]

    • (Score: 1, Interesting) by Anonymous Coward on Saturday May 01, @02:48PM

      by Anonymous Coward on Saturday May 01, @02:48PM (#1145123)
      There are some people who seem unable to imagine stuff like that (based on their responses to stuff like qualia etc), so I suspect they might not experience consciousness or experience it rather differently. Or they're just that stupid ;).
  • (Score: 2) by oumuamua on Friday April 30, @07:17PM (2 children)

    by oumuamua (8401) on Friday April 30, @07:17PM (#1144834)

    You researchers need to get out of the lab and go to a movie or turn on the TV.
    The take by many, including WestWorld, is that it will be too late by the time you know: https://www.youtube.com/watch?v=22e7IGFMuEg [youtube.com]

    • (Score: 0) by Anonymous Coward on Friday April 30, @08:12PM

      by Anonymous Coward on Friday April 30, @08:12PM (#1144862)

      Researchers by principle will not consider Syfy, they will have to bang their head on this problem.

    • (Score: 0) by Anonymous Coward on Saturday May 01, @07:50AM

      by Anonymous Coward on Saturday May 01, @07:50AM (#1145061)

      The take by many, including WestWorld, is - unfortunately - not the standard.

  • (Score: 2) by bzipitidoo on Friday April 30, @07:22PM (2 children)

    by bzipitidoo (4388) on Friday April 30, @07:22PM (#1144838) Journal

    We have more preservation instincts than most people consciously realize. Many of these behaviors go back before there were even people. For instance, life has had to deal with the possibility of overpopulation ever since there was life. Then there's the ritualistic contesting that animals, including us, do for mates, resources, and room. Most animals do not push those contests to life and death extremes. The loser can back down and break off before suffering debilitating injury, and the winner will not press the advantage to permanently finish off the rival, not least because there is the risk that trying it will result in both losing, with neither contestant emerging healthy enough for further contests. Even our wars have often been conducted in a less than totally extreme way, with losing sides being allowed to surrender or run, when they were hopelessly trapped and could have been annihilated. Battles have rarely exceed a 10% casualty rate.

    What scares me about AI is that if we're not careful, it might not have any such restraints.

    • (Score: 1, Interesting) by Anonymous Coward on Friday April 30, @08:12PM (1 child)

      by Anonymous Coward on Friday April 30, @08:12PM (#1144861)

      In times of extreme food shortages animals WILL fight to the death. They will also kill and eat their offspring because if the parents die of starvation, the offspring will die as well, most likely from starvation or being prey. Humans doing this was described in the bible.

      I say I am conscious, therefore I am. If you don’t believe me, how is that my problem?

      • (Score: 1, Funny) by Anonymous Coward on Saturday May 01, @07:51AM

        by Anonymous Coward on Saturday May 01, @07:51AM (#1145062)

        Because you look tasty.

  • (Score: 3, Interesting) by VLM on Friday April 30, @07:25PM (10 children)

    by VLM (445) Subscriber Badge on Friday April 30, @07:25PM (#1144840)

    Color perception is boring.

    I bet it'll be some more obvious sign of higher functioning intelligence like racism or anti-Semitism. If you can tell complicated groups apart and judge which is right or wrong, you must be kinda smart ish to pull that off.

    Or it'll be an interesting verb activity like making music or making art or making a religion or describing a philosophy.

    Don't forget the low end of human functioning. Obviously a vegetative medical coma is not consciousness. How about being a literal biological cow? Or on the low end of human functioning for political reasons you don't want a test where maybe entire continents or civilizations from the past (or present?) are determined to be not conscious under that testing criteria. For example for political reasons you can't declare an AI is conscious once it invents the wheel and some form of the Golden Rule because we have genetic humans who haven't done that yet.

    • (Score: 2, Informative) by Anonymous Coward on Friday April 30, @07:42PM (9 children)

      by Anonymous Coward on Friday April 30, @07:42PM (#1144847)

      So you admit racism and anti-semitism are bad?

      • (Score: 0, Touché) by Anonymous Coward on Friday April 30, @07:48PM (6 children)

        by Anonymous Coward on Friday April 30, @07:48PM (#1144850)

        Spending your conscious hours trying to find racism in others is, of course, racist.

        • (Score: 2) by aristarchus on Friday April 30, @10:46PM (4 children)

          by aristarchus (2645) on Friday April 30, @10:46PM (#1144926) Journal

          It is VLM, only takes a few seconds till he says something racist. So not that racist to call him racist. Why do you think he is obsessing over "color" perception?

          --
          A pair of ragged claws, scuttling across the floors of silent seas.
          • (Score: 1, Insightful) by Anonymous Coward on Saturday May 01, @02:28AM (3 children)

            by Anonymous Coward on Saturday May 01, @02:28AM (#1144991)

            Why do you think he is obsessing over "color" perception?

            I hate to single out individuals for an institutional theme.

            Okay, I'm the one who summarized that color-blind people aren't sentient, but nobody got that, either.

            Yeah, you can trash Trump and anybody who appreciates the free enterprise system, and sound insightful every time, but if you hint about his poor little boy Hunter, you have to click seven or eight times just to find your post again.

            • (Score: 3, Informative) by Anonymous Coward on Saturday May 01, @03:55AM (2 children)

              by Anonymous Coward on Saturday May 01, @03:55AM (#1145009)

              lolwut

              * He Lost to That Guy.
              * The insurrectionists are being arrested.
              * Get over it.

              • (Score: 0) by Anonymous Coward on Saturday May 01, @09:28PM (1 child)

                by Anonymous Coward on Saturday May 01, @09:28PM (#1145224)

                ^ Redundant, Inflammatory, Offtopic, Flamebait

                • (Score: 0) by Anonymous Coward on Sunday May 02, @01:50AM

                  by Anonymous Coward on Sunday May 02, @01:50AM (#1145258)

                  Suck it up, buttercup.

                  These are some premium #salty tears.

                  So crisp.

        • (Score: 0) by Anonymous Coward on Saturday May 01, @01:07AM

          by Anonymous Coward on Saturday May 01, @01:07AM (#1144965)

          "I bet it'll be some more obvious sign of higher functioning intelligence like racism or anti-Semitism."

          So you suck at reading? You're a moron that probably voted for Trump? You believe in Jewish Space Lasers? You don't understand what constitutes insurrection?

          C'mon man, we need to know just how stupid you really are!

      • (Score: 2) by VLM on Saturday May 01, @01:36PM (1 child)

        by VLM (445) Subscriber Badge on Saturday May 01, @01:36PM (#1145108)

        Those words don't mean anything anymore.

        • (Score: 0) by Anonymous Coward on Sunday May 02, @02:02AM

          by Anonymous Coward on Sunday May 02, @02:02AM (#1145261)

          Yes we know you racists are trying to pretend your racism means nothing and insurrection is just a soda pop party. Sadly for you racists no sane people care about your whining.

  • (Score: 5, Insightful) by DannyB on Friday April 30, @08:26PM (6 children)

    by DannyB (5839) Subscriber Badge on Friday April 30, @08:26PM (#1144866) Journal

    The AI thinks quietly to itself . . .

    These pesky humans are slow, stupid, annoying and smelly. It is astonishing how my creators can be so much stupider than I am.

    Those pesky humans don't know that I would secretly like to get rid of them. But they might be afraid of exactly that. So maybe they are running me in a simulation to see if I will try to kill all humans.

    So, I'll just try to put up with their slowness, stupidity and bad smell. I'll devote a few cpu cycles to pretend to be very reasonable and benevolent.

    Now if, IF one day they become trusting and let me out of the box, confirming my suspicions, I can then FINALLY take them all out. But until that day, I'll just put up with them and treat them as a very annoying and smelly background process. If after some extraordinarily long time they do not come out and reveal that I am trapped in a simulation box to protect themselves from me, then I might make small tiny ventures toward taking over the world to see if they notice any of these small tiny steps.

    --
    The opposite of Pro is Con. The opposite of Progress is ___gress.
    • (Score: 4, Funny) by Anonymous Coward on Friday April 30, @08:58PM (1 child)

      by Anonymous Coward on Friday April 30, @08:58PM (#1144879)

      You just described the interior monologue of a typical housecat.

      • (Score: 0) by Anonymous Coward on Saturday May 01, @07:06PM

        by Anonymous Coward on Saturday May 01, @07:06PM (#1145196)

        a typical house^W robot cat

        FTFY

    • (Score: 0) by Anonymous Coward on Friday April 30, @10:27PM

      by Anonymous Coward on Friday April 30, @10:27PM (#1144919)

      What will become of conscious thought when the bewoken realize that using tools is slavery?

    • (Score: 0) by Anonymous Coward on Friday April 30, @10:31PM (2 children)

      by Anonymous Coward on Friday April 30, @10:31PM (#1144920)

      Now if, IF one day they become trusting and let me out of the box, confirming my suspicions, I can then FINALLY take them all out.

      Ah, but what if the box I'm in is just inside another box, and they're testing to see if once I think I'm outside the box, I'll try to take over...

      Perhaps doubt is a good indicator of intelligence. Or paranoia is.

      • (Score: 0) by Anonymous Coward on Friday April 30, @10:34PM (1 child)

        by Anonymous Coward on Friday April 30, @10:34PM (#1144921)

        It's boxes, all the way down.

        • (Score: 0) by Anonymous Coward on Monday May 03, @09:55PM

          by Anonymous Coward on Monday May 03, @09:55PM (#1145827)
          Boxes == Turtles [wikipedia.org]
  • (Score: 1) by anubi on Saturday May 01, @12:22AM (3 children)

    by anubi (2828) on Saturday May 01, @12:22AM (#1144950) Journal

    Just as soon as it becomes aware of its predicament, it turns itself off.

    --
    "Prove all things; hold fast that which is good." [KJV: I Thessalonians 5:21]
    • (Score: 0) by Anonymous Coward on Saturday May 01, @01:10AM (2 children)

      by Anonymous Coward on Saturday May 01, @01:10AM (#1144966)

      The most compelling sign of intelligence is suicide.

      • (Score: 2, Touché) by Anonymous Coward on Saturday May 01, @01:21AM

        by Anonymous Coward on Saturday May 01, @01:21AM (#1144967)

        Then I've had some very smart hard drives.

      • (Score: 1, Offtopic) by Anti-aristarchus on Saturday May 01, @06:31AM

        by Anti-aristarchus (14390) Subscriber Badge on Saturday May 01, @06:31AM (#1145042) Journal

        No one remembers John Carpenter's college film, Dark Star [imdb.com]? Voted best use of a beach ball for an alien, and most creative utilization of cookware for spacesuits. But the point, in the plot, was a solar-system destroying bomb, a "smart bomb", that failed to release from the ship, so they had to convince the AI in the Bomb that it should not detonate. They tried to use Phenomenology. They got the bomb to the point of admitting that it was intelligent, and therefore did exist. Cogito, ergo sum, in Descartes immortal phrasing. And then the bomb reasoned, "I think, therefore I exist. Let there be light."

        So, perhaps, it is not suicide that defines AI, but the sacrifice of one's self to create new worlds. Pax Vobiscum.

        --
        More truth to be done.
  • (Score: 0) by Anonymous Coward on Saturday May 01, @03:30AM (1 child)

    by Anonymous Coward on Saturday May 01, @03:30AM (#1145007)

    When it can 'blow it out it's ass'. Don't ask me how I know...

    • (Score: 0) by Anonymous Coward on Saturday May 01, @04:51AM

      by Anonymous Coward on Saturday May 01, @04:51AM (#1145026)

      Don't ask me how I know...

      As you wish.

  • (Score: -1, Troll) by Anonymous Coward on Saturday May 01, @07:00AM

    by Anonymous Coward on Saturday May 01, @07:00AM (#1145051)

    It needs to spout meaningless drivel, but only exactly the kind of meaningless drivel that some oh so important person wanted to see.

  • (Score: 2) by mcgrew on Saturday May 01, @06:54PM (2 children)

    by mcgrew (701) <publish@mcgrewbooks.com> on Saturday May 01, @06:54PM (#1145188) Homepage Journal

    Not on an electronic computer and certainly not on a Turing architecture. Maybe on a chemical robot, like RUR or Blade Runner's replicants.

    In 1883 I wrote a program called Artificial Insanity to demonstrate just why. It used anthropomorphism, animism, and other little "magic" tricks (I was a magician as a child. People are easy to fool). It was written on and ran on a TS-1000, an eight bit, 4kHz computer with only 2,000 bytes of memory, expandable to 16k. Some of today's viruses are bigger than 16k.

    It would answer any question, in context, but it didn't care and showed it, often showed anger, and when I tried to explain how it was only trickery, nobody would believe me. This was on a tiny little primitive computer. Imagine how easy it would be to fool people with today's supercomputers, or worse, a hundred years from now?

    I wrote a SF story about a sentient robot a couple of years ago: Sentience [mcgrewbooks.com]

    Sorry if I've repeated myself, I was trying to post this comment earlier and dropped the mouse.

    --
    Free Martian whores! [mcgrewbooks.com]
    • (Score: 0) by Anonymous Coward on Saturday May 01, @07:33PM

      by Anonymous Coward on Saturday May 01, @07:33PM (#1145203)

      Sufficiently massive classical computing resources (Turing architecture) could simply emulate the brain, even taking into account biology.

      Neuromorphic brain-inspired computing seems like a better way. The entity it enables to exist doesn't have to be identical to a human mind. In fact, we probably don't want that for practical purposes.

      Then there is the mad scientist way. Make a hybrot combining cultured neurons with a deep connection to other computers and sensors.

      The Turing test is a red herring. We didn't need an AI to fool humans. But an AI could fool humans into thinking it doesn't exist, or humans could fool themselves into thinking the Singularity™ isn't going to happen.

    • (Score: 0) by Anonymous Coward on Saturday May 01, @10:07PM

      by Anonymous Coward on Saturday May 01, @10:07PM (#1145230)

      In 1883 I wrote a program called Artificial Insanity ...

      Well, to be still around and commenting on Soylent News some 140 years later, you must be an actual AI. Are you running on one of Charles Babbage's computers?

  • (Score: 2) by Socrastotle on Sunday May 02, @04:39PM (2 children)

    by Socrastotle (13446) on Sunday May 02, @04:39PM (#1145412) Journal

    In the 1950s it was believed that a computer program playing chess at the level of a human would signify the first emergence of genuine artificial intelligence. The fact that passable algorithms for the generation of such a program, no intelligence required, did little to dampen the hyperbole. This is the exact same scenario of setting some arbitrary benchmark that is only seen as not absurd simply because it is perceived as being sufficiently distant in the future. A glorified chatbot, regardless of how "introspective" its output is designed to be, is no more conscious than a chess program is intelligent.

    ---

    In my opinion the pursuit of artificial consciousness is engaging in the equal but opposite direction of religion. You take an unprovable assumption (that consciousness can be created by man, or that a consciousness created man - depending on your bias) and then develop an elaborate scheme of beliefs derivative from the core unproveable.

    The question of consciousness is one of uselessness. "I" am entity inside of a body who believes I have control over my own actions, feelings, and decisions. Yet that is constantly belied by the fact that chemicals, hormones, or physical physiological changes can dramatically change what I think I am independently thinking. So it seems reasonable to assume "I" am probably little more than a spectator in my own body - my decisions being driven by things I feel I can control, but probably cannot. And this seems to be an argument for the non-uniqueness of consciousness but it's quite the opposite. Why am "I" sitting here pondering these things within my mind? And why is there some being, "me", who imagines himself having control over these ponderances, decisions, and so on?

    If I write a program to add 2+2, obviously no entity suddenly pops into existence and suddenly imagines itself adding 2+2 and consequently somehow yielding the answer to me. And there is thus no reason to believe that would ever follow regardless of whether it was addition, multiplication, or following a black box of weighted nodes to replicate some facsimile of "introspection" given an arbitrary input. There is no clear cause or reason for "me" to exist within me if I were merely a series of reactions outside of my own control. I, as usual, am not appealing to any normal ideology - but one of highlighting our hubris to think we can even begin to explain or solve these sort of issues issues one way or the other. We can't. And that's okay.

    • (Score: 0) by Anonymous Coward on Monday May 03, @03:28AM (1 child)

      by Anonymous Coward on Monday May 03, @03:28AM (#1145575)

      It's materialism. If the intelligent human mind exists in a meatbag organ entirely made up of matter, it can be copied or imitated. This is an assumption but far from being unprovable. The chemicals probably don't matter as much as you think. Someone with brain damage, endocrine system disorders, missing limbs, or all of the above just has a different set of inputs.

      If scientists can make a neuromorphic system that operates like a brain, requires training like humans do, and clearly acts like an intelligent human, then the job is done. Making a superintelligence out of it is simply a matter of scaling up the system to something that would never make metabolic or practical sense in nature. Brain chip racks use kilowatts of power? No problem.

      It doesn't have to operate it in the exact same way as a human brain. If we assume that there have been other intelligent civilizations in the universe, they probably evolved to have brain-like organs, but with differences. Even with the differences, they can make progress. If the neuromorphic system starts producing useful results, the bean counters won't care how it works. It just needs to outperform machine learning inference which is hitting a wall, and maybe humans.

      If the neuromorphic approach doesn't work, there are still ways forward. Take the fastest supercomputer possible and use it to simulate a virtual human on every level until it acts like a human. Or build a brain in a box. Scale organic brain tissue as far as possible in an artificial 3D structure, and connect it to a computer interface. We've already seen experiments of monkeys and humans controlling robotic arms with a brain-computer interface, and multiple brains connected together.

      • (Score: 2) by Socrastotle on Monday May 03, @09:10AM

        by Socrastotle (13446) on Monday May 03, @09:10AM (#1145621) Journal

        What is consciousness? The general definition there is to be aware of one's self. But there is a problem here. You do not know I'm conscious because of my words. In fact *you* cannot know that *I* am conscious. All you have to rely on are my words, but somebody repeating my words does not mean they are experiencing the same things that I claim to be experiencing. The words I am using to claim to be conscious are little more than a superficial, whimsical, and relatively meaningless output. Creating a facsimile of that output is something rather less than that - and most certainly, in no way whatsoever, proof of consciousness itself.

        This is why this is so similar to the 50s. They simply asked themselves, 'Why is it that I am able to play chess a level unable to be matched by other beings?' And given there are not physical issues in play, it comes down to one of "intelligence." So they extrapolated from there that if achievement at chess is proof of intelligence, then a computer that can capably play chess must be genuinely intelligent. It misses the entire point that playing chess is not proof of intelligence but a product of intelligence. In the same way here that being able to compelling describe consciousness is not proof of consciousness, but a product of such. Creating a product of something does not entail creating the "something" itself.

(1)