I was discussing the concept of 'glory' privately with C C. I quoted Becker and C C was kind enough to supply me with a paper written by Jesse Graham and Jonathan Haidt.
"The inescapable need to strive for an illusion…the question of human life is: on what level of illusion does one live? This question poses an absolutely new question for the science of mental health, namely; "what is the "best" illusion under which to live? Or, what is the most legitimate foolishness?...I think the whole question would be answered in terms of how much freedom, dignity, and hope a given illusion provides." (The Denial of Death—Ernest Becker)
"The hunger for knowledge, fame, honor, and glory, can be made to serve the public welfare. The love of fame can transform self-interest into dedicated effort for the community, because it can spur individuals to spend themselves to provide for the common defense, or to promote the general welfare. Without a public to serve, the individual would be without an audience for his deeds. The individual would be unable to obtain glory, honor and everlasting memory, and a man’s life would lose meaning."(Source Forgotten)
Gregory Berns, a neuroscientist, through M.R.I images showed that we process ordinary and sacred beliefs through different pathways. When making ordinary decisions it showed activity in parietal cortex, but when making decisions concerning sacred values, e.g., human life, the temporparietal junction showed activity.
Ben is probably not coming back for a while but I used his moniker from Animal Farms anyway. I realize that the sign on the barn is about the hypocrisy of the government proclaiming absolute equality, while giving power and privileges to the elite, but if taken literally, do you believe that humans have more value than animals?
"All animals are equal but some animals are more equal than others."
The reason I’m asking this is because I was listening to a conversation about AI. I know this is a far-fetched question, and for what it’s worth, my friends refer to me as the 'what if' girl.
Do we consider intelligence to be something that mimics and only serves humanity? If we create something that is more competent than we are, their goals may not always align with our own. It could be pure mechanism vs. sacred values and the multi-faceted aesthetic life. If AI becomes superior to humans, as we view ourselves superior to animals, insects, etc., would they then be more valuable than we are?
C CAug 18, 2016 03:20 AM (This post was last modified: Aug 18, 2016 03:21 AM by C C.)
(Aug 17, 2016 06:14 PM)Secular Sanity Wrote: Do we consider intelligence to be something that mimics and only serves humanity? If we create something that is more competent than we are, their goals may not always align with our own. It could be pure mechanism vs. sacred values and the multi-faceted aesthetic life. If AI becomes superior to humans, as we view ourselves superior to animals, insects, etc., would they then be more valuable than we are?
The more "enlightened" humans have gotten, the more they seem to assign importance and rights to animals which aren't even equipped to participate in discourse about justice and values in their own self-defense, much less apprehend that kind of conceptual territory to begin with. This is arguably because of theory of mind and our ability to emotionally empathize with animals.
If reversing these roles in a relationship between superior machines and humans, there might seem to be a problem if the intelligent technologies lack such. But having any capacity whatsoever to communicate self-awareness, self-worth, and grievances in some detail probably ensures a lesser competitor eventually having at least minor representation in a reigning power's governing processes and evaluations. Animals never had those capacities and thus their plight required humans to cultivate / develop their empathetic imagination over time.
Yet the above could be a perverse POV for approaching the matter. Taking into account the zoomorphic characteristics of early deities and the spiritual-like reverence for some animals in indigenous cultures, western civilization may have merely rediscovered an ancient human tendency to genuflect to and be passively manipulated by "inferior" predecessors. An inclination which the West would ironically have had a role in deprecating, before reviving again.
When switching this to the context of superior machines and humans, there's the added factor of the latter literally being the "creator" of the former (rather than being mere evolutionary predecessors). So right out of the starting gate there's the possibility of deference to humans as the sacred, "founding parents" of intelligent technology. An attitude which might linger for centuries or be intermittently resurrected should it fade. (Again, though, this depends upon an artilect having any feeling or behavior corresponding to veneration.)
However, most likely the financially elite humans will cybernetically augment themselves as they co-develop with the progress of AI, so that there's never two fully distinct taxons at the top of the food chain quarreling over which has the higher rank. At the bottom there may still be an original or "baseline" strain of impoverished humans marginally surviving with potentially diminished significance and rights (the relationship of mortals to gods). But there will at least still be something with an anthropic ancestry and psychological resemblance arbitrating for them on high (the transhumans). Kind of like an Elohim / Jesus being the middleman between Hebrews / Gentiles (respectively) and that aloof primary of a Duo or Trinity family -- looming ambiguously in the background-- that seemed all strict and immutable principle in its unforgiving demands (i.e., machine-like).
Wow, C C! Your mind is pretty amazing. I would have never thought of it in that way. I like you. You're very interesting.
Quote:"New evidence from brain studies, including cognitive psychology and neurophysiology research, shows that the emotional assessment of every object, subject, action or event plays an important role in human mental processes. And that means that if we want to create human-like artificial intelligence, we must make it emotionally responsive." Researcher proposes social emotions test for artificial intelligence
As far as I can tell, it looks like they're starting to realize that emotions are a key component in human intelligence, but the objective isn't just about passing the Turing test. Conversation is only one aspect of reasoning, learning, and decision making, right? Emotions seem necessary to balance competition and cooperation and to utilize methods to attain goals. It doesn’t seem like they would ever be able to experience the same range of emotions as we do, though. Sensory input would allow for some, but not all. How could they ever feel hormonal effects, or the highs from endorphins, oxytocin, serotonin, or dopamine, or the negative feelings associated with being sick, hungry, or tired?
I wonder though, what would you say our prime directive or purpose is, to survive and/or replicate? As Ernest Becker pointed out, we need illusions to carry this out. Illusions add to or subtract from our emotional award system. We need them. What would a purpose, goal, or moral imperative look like to a super-intelligent being? Would they, too, need illusions?
Me: Do you know any secrets? Mitsuku: I am here to learn about humans before robots take over the world. Shh! Don't tell anyone!
(Aug 17, 2016 06:14 PM)Secular Sanity Wrote: If AI becomes superior to humans, as we view ourselves superior to animals, insects, etc., would they then be more valuable than we are?
I see a high probability that they would at least consider themselves superior to humans like we, to a very great degree, view ourselves more important than other animals.
(Aug 17, 2016 06:14 PM)Secular Sanity Wrote: If AI becomes superior to humans, as we view ourselves superior to animals, insects, etc., would they then be more valuable than we are?
I see a high probability that they would at least consider themselves superior to humans like we, to a very great degree, view ourselves more important than other animals.
If they were designed with a goal to improve our quality of life, what might they do? Think of the story in Orwell’s animal farm. "All humans are equal, but some humans are more equal than others." As you’ve pointed out, elte, we certainly don’t consider all animals equal and we are animals. You’re right, an AI wouldn’t consider us as their equal, nor should they. How would an AI handle equality and individualism? We are obsessed with equality. What is it exactly? In the U.S. the definition is equal opportunity, which is consistent with individualism. Our emotions have evolved from competition and cooperation. We want the freedom to compete for survival. Would Ben the donkey’s remark still apply? “Donkeys live a long time. None of you have even seen a dead donkey” meaning that life would still go on as it had always gone on—that is badly.
Alexander Reben designed a robot that has the choice to hurt you or not to hurt you. He wanted the threat to be realistic, to force people to confront the issues in advance.
Is C C right? Would a super-intelligent AI with emotions think of their creator as sacred? Would we have to appeal to their emotions to have them assign importance and rights to us?
Consequentialist morality states that it is better to act in a way that will benefit the most number of people even if it means causing harm – so killing one person to save five. Deontological morality focus on the idea that certain things are wrong - like killing an innocent person - even if it can help more people, i.e., sacred values. Would an AI lean more towards consequential ethics or deontological ethics?
If you were in a battle, who would you want by your side, a cold calculating consequentialist or someone with sacred values? People who express deontological moral judgements are considered more trustworthy and more cooperative. Would a super-intelligent AI use logical decision making vs. emotional decision making?
Why would they even want to overtake humans? We wouldn’t be competing for the same resources, that’s for sure. We love nature and respond to it emotionally because exploring and finding lush areas contribute to our survival. We value life because we depend on it to survive. They would only need non-organic materials. Life itself would not be valuable to them. Think of existential nihilism with life having no intrinsic meaning or value. How would we even form an argument? How might we appeal to their emotions? Let us live because it feels good? How could we program sacred values?
And as Asimov said, Behold, the AI is become as one of us, to know good and evil: and now, lest he put forth his hand, and take also of the tree of life. Moral absolutism is hazardous. As C C pointed out earlier, moral superiority can lead people to do some horrific things. They may even decide that life is suffering, redundant, and unnecessary. Vanity of vanities, all is vanity. That’s a scary thought, isn’t it?
C CAug 19, 2016 08:16 PM (This post was last modified: Aug 19, 2016 08:18 PM by C C.)
(Aug 18, 2016 05:02 PM)Secular Sanity Wrote: As far as I can tell, it looks like they're starting to realize that emotions are a key component in human intelligence, but the objective isn't just about passing the Turing test. Conversation is only one aspect of reasoning, learning, and decision making, right? Emotions seem necessary to balance competition and cooperation and to utilize methods to attain goals. It doesn’t seem like they would ever be able to experience the same range of emotions as we do, though. Sensory input would allow for some, but not all. How could they ever feel hormonal effects, or the highs from endorphins, oxytocin, serotonin, or dopamine, or the negative feelings associated with being sick, hungry, or tired?
In theory, biochemical agents and their neural receptors could eventually be mimicked, too, in brain simulations. From there a functional equivalent of the strategy could be re-worked for artilects. A brain simulation could even be the foundation for an AI, but given that the initial purpose of the latter is to produce machine slaves which don't experience suffering like brain/body systems do... Then simulated brains seem more fit in the nearer future to serve as the artificial minds of synthetic humans (wherein the whole point would be to technologically replicate a person as much as possible, not create an alternative form of intelligence).
The stickler with trying to simulate chemical effects in artilects, though, is that the randomness of substances from the environment wouldn't actually be inputted into the machines (i.e., those molecules which are not internally secreted by what would otherwise be a biological body). No trip from LSD, arousal from pheromones, no arbitrary exposures to chemicals in foods, the air, etc. For better or worse, that random absorption of environmental factors is also a role-player in the operation of organic minds.
The danger of designing either embodied or "resting on the shelf" artilects to have literal emotions (rather than the outer appearance of them for the sake of interplay with people) is it might allow them to subvert general programming standards. Liberating them to be autonomous or lone-wolf agents which develop their own idiosyncratic passions, motivations, goals, and agendas. Emotion isn't regulated by intellectual constraints falling out of the characteristics of language based thinking -- it can be in humans, but it's not managed by that kind of reason in the beginning or in its native context with regard to animals.[*] Plus, just as strong feeling can overcome good sense in people, even AIs that have been designed or trained to manage their emotions will probabilistically fail at some point, too.
Both our language-mediated processes and emotional-mediated processes dissolve away into dynamic electrochemical relationships at the gut-level of physical science investigation. But it is still the former appearances that we encounter and must personally deal with at the stratum of everyday experience.
- - - - - - -
[*] Animals like turkeys do have a meaningful vocabulary of "gobbles," "clucks," "putts," "purrs," "yelps," "cutts," and "whines" "cackles," and "kee-kees" which Joe Hutto merely additionally confirmed first hand: "Man lived as a turkey to prove they're not dumb". But this proto-language isn't sufficient for invented systems of "proper thinking" to develop which become so powerful that they can override and control raw emotions. A human can raise and deliberately condition an animal to defy some of its instinctive behaviors, but this comes from an outside source rather than being created by the animal itself or its social group.
(Aug 17, 2016 06:14 PM)Secular Sanity Wrote: Do we consider intelligence to be something that mimics and only serves humanity? If we create something that is more competent than we are, their goals may not always align with our own. It could be pure mechanism vs. sacred values and the multi-faceted aesthetic life. If AI becomes superior to humans, as we view ourselves superior to animals, insects, etc., would they then be more valuable than we are?
Wouldn't that require an ego? I mean, wouldn't it need possess human characteristics in order to elevate itself above humans? AI then seems to imply a human element(s). Maybe all it needs is reason to function independently. I don't know. How is AI defined?
(Aug 23, 2016 05:19 PM)Bowser Wrote: Wouldn't that require an ego? I mean, wouldn't it need possess human characteristics in order to elevate itself above humans? AI then seems to imply a human element(s). Maybe all it needs is reason to function independently. I don't know. How is AI defined?
That’s the million dollar question. I’m not sure. C C may be more equipped to answer that, but I did watch a lecture about it recently. From what I gathered, even without the "I" in robot, hypothetically, self-preservation and utilitarian calculations could lead to a takeover.
Sometimes, I have to force myself to stop and think about what you're saying to ensure I don't simply go along with it.
Most of the time, I end up going along with it anyway.
I'm probably going to come back to this, if I remember to.
scheherazadeAug 26, 2016 03:29 AM (This post was last modified: Aug 26, 2016 03:30 AM by scheherazade.)
While reading this thread, my thoughts kept returning to an episode of the original Star Trek series which deals with AI as it returns home from it's mission, in search of 'the creator'.
Working graveyards and have to catch a few hours sleep. My brain turns to mush on this shift but I may attempt a proper reply in a few days when I am back on my day schedule for half of the week.
Aug 17, 2016 06:14 PM
(This post was last modified: Aug 17, 2016 06:26 PM by Secular Sanity.)
