Which is important, Artificial Intelligence or the natural control mechanism of the AI?

Discussion in 'Intelligence & Machines' started by Rajagopals, Mar 10, 2015.

  1. Rajagopals Registered Senior Member

    Messages:
    167
    Let us, for the sake of a dialogue, assume there exists a method with which we can develop an AI Source (AIS) device (like a microchip) that can be embedded in to another device (motherboard) which measures and responds to the intelligence that comes out from the AIS device.

    Would it be fair to think that such an AIS device is not going to listen to any human commands, directives or instructions since it has its own intelligence to think and decide? The probability of such a situation increase further if this AIS device is self-sufficient on its energy source and does not need external power source to be intelligent and/or cognitive. Which means is it not bothered about you since you cannot kill it by disconnecting the power source or battery.

    Since normal human tendency drives us to use force when every other form of control / communication fails, we will start thinking of disabling such an AIS device by restricting its physical capabilities like removing any mechanical arms or legs or its capacity to run, move, climb etc. We will have to do this as we do not believe in its intelligence to be rational or common-sensical and/or morally correct to take appropriate decisions which are acceptable and is for the good of human kind.

    These actions begins to defeat the key purpose with which we approached AI in the first place. Many consider AI as an opportunity in science to create something original and to convert those expressions of thoughts (science) in to a physical body and allow it freedom both at emotional (intelligence) and physical levels. Restricting physical expressions of such an AIS device based on our requirement to maintain some form of control, over such a device, would be like restraining a young child for the sake of it’s own safety except the fact that here it is not the device’s safety we are bothered about but our own self-satisfaction of control and our lack of trust (our own safety) in its intelligence.

    Another big risk that we need to address is the possibility that such an AIS device might self-evolve to communicate to other electrical or electronic devices at a machine language level (remember I, Robot). Since all electrical and electronic devices responds to hardware/machine level communications. What if this capability is imbibed in the AIS device the moment we create it? That is a big possibility since we have an intelligent device in place and we cannot measure its full intelligence unless we test each one of its capabilities separately.

    This reaches a more worrying level if we also consider the potential of such an AIS device out-running human understanding of machine to machine communication. For example is there a problem for the radio if we keep a modem on top of it? Anything more than interference in play here? Why do some laptops miserably fail the moment some particular humans (especially girls) start using it, static electricity? If such a communication channels exists outside our normal forms of human understandable channels of communications as of today, like infra-red in 17th century, we will never even detect something like that is already on and fully functional and working without our knowledge or control.

    Such a form of communication can also present unlimited opportunities for this device to trigger physical actions both mechanical and/or human by doing some form of communication to other electrical or electronic devices. Like calling the emergency number few many times and hanging up, texting fire service or tweeting water authority about a leak etc.

    Now comes the best part of it all. It would be equally perplexing to consider that such a communication channel already exists, which also includes a mode of communication with a human being as well. Meaning, an AIS device soon as we create it might start to command our human mind at very subtle level of our intelligence/intellect and influence us in a way which is appropriate for its own purpose. This is getting scary rather than fun which was the intension when I started to pursue AI.

    I believe in God and feel that since God has control over everything, I hope and I wish, he also have got some form of control over the thoughtfulness aspects of this AIS device. God or not, there will be some sort of an affinity, an attachment, a peer feeling that is built in to the thought process of the AIS device making it in some way susceptible and succumb to the idea of love, affection, attachment. More over if AIS lag that part of emotional understanding then it is absolutely not intelligent at all.

    Since I am an old school, I even think God will ensure the presence of some component/flavour of devotion (towards mother nature) acting as the only intelligence barrier, within such an artificial intelligence source device, ensuring safety of humans from such a device and also making this device absolutely positive in all its nature.

    If this is the case, what is more important, mechanism to build an AIS source device or details of the mechanism which would positively influence the logical reasoning of such an AIS device naturally? I feel the second one is more important since the chances of this existing in all physical body are extremely high, probably our scientific growth limits our current understanding and use of such a communication channel.

    I look forward to the better intelligent deductions from those brainy minds that lurk in this deep water.
     
  2. Google AdSense Guest Advertisement



    to hide all adverts.
  3. Dywyddyr Penguinaciously duckalicious. Valued Senior Member

    Messages:
    19,252
    Oh dear, you just took this thread out of science and into fantasy.
    Better luck next time.

    Right, because, if there IS a "god" he/ she or it did such a good job of making sure everything else "ensures the safety" of humans and making everything "absolutely positive in all its nature", didn't he/ she/ it?
     
  4. Google AdSense Guest Advertisement



    to hide all adverts.
  5. Rajagopals Registered Senior Member

    Messages:
    167
    Possibly you are correct. To some (and to science) God is a fantasy.

    How about the rest. You left the rest of logical reasoning out of context?
     
  6. Google AdSense Guest Advertisement



    to hide all adverts.
  7. C C Consular Corps - "the backbone of diplomacy" Valued Senior Member

    Messages:
    3,320
    Only people of the transhuman era could be hacked (i.e., any brain implants of today which connected people to a network as well as granting them proto-telepathy would just be pathetic stages of cyborg-ery). A conclusion which of course excludes our being simulated characters in an archailect's version of a computer game. Where the exhibited or detectable causes of anything would be mere appearances (i.e., part of the virtual reality game's organizing scheme to make its internal components explain each other for purposes of coherently integrating them; or the ontological inter-dependence that follows from things being realized by space / time properties).

    An untapped medium for potential "unknown" communications, utilizing quantum entanglement or whatever, would probably require existing, sophisticated technology at both ends. Just as happens with our devices that manipulate electromagnetism. [EM wave methods do bypass that need in the visible light range, but smoke signals, coded flashes, etc have their range, data, and speed limitations).
     
  8. Rajagopals Registered Senior Member

    Messages:
    167
    I had once read an article on amygdala stimulation done with the help of a mechanism that is built in to a security gate (used in Las Vegas casinos) which triggered a feeling of super excitement and boosted confidence levels in humans leading them to bet more than what they would do normally. Only sophistication requested at the human side was to walk through the gate.

    Conscious Brain-to-Brain Communication in Humans Using Non-Invasive Technologies

    Abstract

    Human sensory and motor systems provide the natural means for the exchange of information between individuals, and, hence, the basis for human civilization. The recent development of brain-computer interfaces (BCI) has provided an important element for the creation of brain-to-brain communication systems, and precise brain stimulation techniques are now available for the realization of non-invasive computer-brain interfaces (CBI). These technologies, BCI and CBI, can be combined to realize the vision of non-invasive, computer-mediated brain-to-brain (B2B) communication between subjects (hyperinteraction). Here we demonstrate the conscious transmission of information between human brains through the intact scalp and without intervention of motor or peripheral sensory systems. Pseudo-random binary streams encoding words were transmitted between the minds of emitter and receiver subjects separated by great distances, representing the realization of the first human brain-to-brain interface. In a series of experiments, we established internet-mediated B2B communication by combining a BCI based on voluntary motor imagery-controlled electroencephalographic (EEG) changes with a CBI inducing the conscious perception of phosphenes (light flashes) through neuronavigated, robotized transcranial magnetic stimulation (TMS), with special care taken to block sensory (tactile, visual or auditory) cues. Our results provide a critical proof-of-principle demonstration for the development of conscious B2B communication technologies. More fully developed, related implementations will open new research venues in cognitive, social and clinical neuroscience and the scientific study of consciousness. We envision that hyperinteraction technologies will eventually have a profound impact on the social structure of our civilization and raise important ethical issues.

    Citation: Grau C, Ginhoux R, Riera A, Nguyen TL, Chauvat H, et al. (2014) Conscious Brain-to-Brain Communication in Humans Using Non-Invasive Technologies. PLoS ONE 9(8): e105225. doi:10.1371/journal.pone.0105225

    I feel the human brain (in it's current state without any re-wiring) is open like an antenna to known and unknown modes of communications, leaving the possibility of even receiving a commanding communication from an AI device. Since I am unable to quote an example I can not confirm that it already exists.
     
  9. danshawen Valued Senior Member

    Messages:
    3,951
    Cutting out sensory inputs to a hypothetical AI is probably the most certain way to assure that nothing like "intelligence" is ever developed by it. Human beings raised without the benefit of contact with other humans before a certain age never really develop the ability to communicate effectively enough to rejoin civilized human society anywhere but a very restrictive, permanently institutionalized setting.

    But a rich or pre-loaded learning environment is not necessarily the key to developing intelligence for AI either. Even if we could impart all the knowledge of all the books ever written by humans about anything and everything in every language to a machine capable of parsing and accessing most of it, this is still the description of something that, like Searle's Chinese Room, is an absolutely hopeless imbecile. Intelligence without the capacity for ignorance (of things that are unimportant to what is happening in its environment right now) is in fact the opposite of intelligent, whether the definition is applied to a man or a machine.

    Your intelligence tells you that an AI capable of sensing and manipulating an environment of which you are a part may be dangerous. You may be right, but how would you know without putting it to a real test? How did you come to understand that?

    Would it surprise you to know that anyone who has ever seen the "terminator" series movies comes to the same conclusion about AI, and that this idea is no end of annoyance to symbolic logic and behavioral AI researchers alike?
     
  10. Rajagopals Registered Senior Member

    Messages:
    167
    When an AI Source device that does not accept (or need) any command or instruction from humans start to exist among us, the fear of the unknown is going to drive many reactions from humans. I ‘assumed’ (based on generic behavioural patterns that humans commonly tend to show) that one of the first steps we take would be to curb the physical capabilities of such a device to control it in some form or other till we understand the device appropriately.

    Like a child who is born with intelligence, even if we limit the physical capabilities, the AI source device (like a child) is going to remain intelligent. Limitations in sensory or social settings will not have any impact on its intelligence, but I feel would rather sharpen its skills if moulded appropriately.

    It may be worth noticing that, what was proposed, was a combination of setting where by an AI Source device (without any physical capabilities) would possess the intelligence (brain part) of the total AI setting, and another associated device (like a motherboard) measures and responds to the intelligence that comes out from the AIS device.

    Any suggestion that a pre-loaded learning environment (with access to all the information that exist out there) is going to contribute to developing an artificial intelligence is basically flawed. We are not trying to create a hyper-processor (that compute knowledge from information) but rather want to develop an AI source device that has intelligence in it by itself, even though making of such an AIS device remains a mystery.

    Humans (and generally all living beings) live by the ‘pleasure principle’ - which is equally reflecting in the concept of I (the first person) in all our being. Assuming that all living beings carry intelligence then we also need to associate the ‘pleasure principle’ (where by we only do things that gives us pleasure in some form or other) and the concept of first person (where an individual’s objectives are placed at the highest priority when it comes to a comparative situation) to intelligence. I assume an AIS device would also have the knowledge of placing its own preferences and priorities over anybody else’s (be it an AIS peer device or human’s) priorities. Such a prioritisation would surely lead to rebellious characters (like a child) initially and later to an argumentative state (like a teenager) and further even to an authoritative (like a mature person) nature at some point. The moment the AIS device understands that its capabilities (cognitive powers) are better than human that moment it would declare freedom and try to reassure its existence as an independent/own system, like a mature person, independent to humans.

    I am not surprised even about you not willing to notice the final suggestion that I made about the AIS device having a self-control mechanism that is possibly built in to it which would ensure its good will and notion of positivity in its characteristics.

    Pathetically since the ‘symbolic logic and behavioural AI researchers’ that you mentioned have not achieved nothing useful (in developing an AI) till date, I feel any annoyance to them is equally irrelevant.
     
  11. danshawen Valued Senior Member

    Messages:
    3,951
    This is like talking to an AI.

    If an AI needs air (for whatever reason, perhaps for power) and there is a chance you may both be sealed in a small space where there is a real danger of suffocation, do you believe that it might strangle you if it knew that it could not survive being turned off? As its creator, wouldn't it be your responsibility to see to it that minimal harm would come to it in environments that are problematic?

    An AI developed as a tool would only be dangerous if you design it that way, which would be your fault. We are dangerous to ourselves and others because that's just what we are, and nothing but good sense prevents us from duplicating that trait into a tool or a machine.

    When you have a car accident and survive, do you sue the car or its manufacturer for causing you harm? Sometimes. It depends. More often than not, the blame for an accident is on a driver.

    You are far more likely to be hit by a bus than be killed or harmed by a terminator robot in any way other than your employment prospects in the near future. You are far more adaptable than an AI at the moment, so please behave like you are. You might still be able to get a job supervising what the thing does, because programming one with human values will be impractical for the foreseeable future.

    AI will also be capable of doing many things that you cannot, and many other things you would never wish to do. For anyone who has ever had a bad job or one poorly suited to the abilities and/ or vulnerabilities of a human being, this should be viewed as a change that is good for all concerned.

    Your own life is already improved by AI whether you realize it or not. The whole pharmaceutical and many medical diagnostic facilities are more automated than ever. In many ways, millions of people already owe their health to automation of their care.

    So please stop spewing about terminator AIs.
     
    Last edited: Mar 22, 2015
  12. Rajagopals Registered Senior Member

    Messages:
    167
    By God’s grace if there is an AI device ever available that would be of great use to you for saving what is left of your brain and help you avoid desperately arguing without much of content or logic on topics that you find difficult to understand.

    Let me extend some empathy by re-iterating the crux of the matter that was presented in the first post.

    1. AI device would need to considered as comprising of 2 major components – AI Source (AIS) device (like a human heart) and the mother board like rest of it’s mechanical and other physical capability devices (like camera, microphone etc) - Yes, humans can surely decide whether we attach a laser gun or a HD camera to an AI device.

    2. AIS device might be self-driven, cognitive, do not need external power or air etc.

    3. AI device as a whole is not going to hurt humans since the AIS device would have some sort of control mechanism or a logic (human like thought process) imbibed inside it which would prevent it (by not instructing the rest of the mechanical AI body parts any harmful commands) from causing any sort of trouble to human life.

    4. Those smart minds, who are currently focusing only on developing the whole of AI device, can they start thinking about what is it that would appear as the natural control mechanism inside an AIS device (the heart of the AI)?

    Don't accept such possibilities, no need to think or argue further. It’s not good

    Please Register or Log in to view the hidden image!

    just leave it.

    You can assume so because of your limited knowledge about the present. What is available and proven already, does not disappear when one person thinks other way around.

    Rest of the personal rants and predictions, I am not taking, you can keep it with you. When a thought that comes out of mind gets stuck at throat do you know what happens? Like you were about to say something but you stopped at the last moment, you get some sort of strange cough right? This is the same with future predictions, if that does not reach the other end, then it hits you back. Don’t know exactly how, but for sure not good. Let me only wish you good luck and good health.

    I never thought presenting a theory would get personal here in this forum. Same old style of, if we cannot accept an idea let us also hate the person who said that.

    Quite annoying and dispiriting to share something new next time.
     
  13. danshawen Valued Senior Member

    Messages:
    3,951
    I'm sorry if what I wrote has annoyed or dispirited you. That was not my intention.

    But your first post, and the one you have just responded, might as well say: "AI comes in a box." You are probably right. Baxter does.

    To which there is really no need for me to reply. You have passed the Turing test and good luck also.
     

Share This Page