1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Rogue tensor flow: Is AI always bad?

Discussion in 'STEM-Science Technology Engineering Mathematics' started by hypnos, Sep 11, 2019.

  1. hypnos

    hypnos Moderator Staff Member

    Joined:
    Aug 7, 2018
    Messages:
    1,961
    Likes Received:
    2,370
    Tensor flow is a open source machine learning environment that is programmable and tweakable by the user.

    It is looking to become a major player in machine learning, which as of now is the age we are stepping into.

    The age of intelligent machines.

    Is AI always bad? I think that may be the next most important question facing preppers for the next 20 years.

    But also, can preparedness minded people develop an intelligence that we can trust, that holds our principles as its prime directive?

    The concept of the Rogue tensor flow could gain some traction, as a counter to high powered mainstream AI used against constitutional American society.

    Maybe the concept is useless, but maybe it can warn us, maybe it can help us predict zero day occurrences, market fluctuations and emergencies by disseminating information at a faster pace than any group of humans can, and remain ever vigilant, always.

    Can the power of AI be harnessed for use for us, or is our unfortunate fate tied to its improvements and development?

    https://determined.ai/product/?gcli...W7TE-i7SXL2bYzCOc7gcS5oMLaOFcVyBoCqwAQAvD_BwE
     
    Atlas likes this.
  2. twp

    twp Administrator Staff Member

    Joined:
    Jun 16, 2018
    Messages:
    3,771
    Likes Received:
    4,288
    Resistance Is Futile, You Will Be Assimilated...
    Not in my lifetime (hopefully) but I can see a trend in that direction. Got your chip yet?

    I can't direct an interstellar ship at the speeds needed for transit, so an AI interface may be needed.
     
    Atlas and hypnos like this.
  3. hypnos

    hypnos Moderator Staff Member

    Joined:
    Aug 7, 2018
    Messages:
    1,961
    Likes Received:
    2,370
    I like frito lays kettle chips (black pepper and sea salt...positively devine)

    Not the other chips tho...

    We use this tech all the time. All you need to do Is call a automated phone system and say something that recognition is AI driven.

    Combine that with RFID, that's a whole seperate dimension. But I suppose it does have uses, although I despise it. No one wants to talk to a machine.

    Maybe it's only terrible, it's all terrible, if so, the internet age brought that on. Does that make the internet evil too?
     
    Atlas and twp like this.
  4. twp

    twp Administrator Staff Member

    Joined:
    Jun 16, 2018
    Messages:
    3,771
    Likes Received:
    4,288
    It's all in how it is used.
     
    Atlas and hypnos like this.
  5. WolfBrother

    WolfBrother Active Member

    Joined:
    Jul 7, 2019
    Messages:
    231
    Likes Received:
    542
    WB edited to specific:
    Before he passed on, Asimov wrote a couple of pre-Foundation novels ( logical time line - events happening before the Foundation series time line started) and several post-Foundation novels (logical time line - events happening after the Foundation series time line ended). In one of the stories, he was able to show a way to enable robots bound by the 3 laws to be able to kill humans.

    The Three Laws are:
    1. A robot may not injure a human being or, through inaction, allow a human being to come to harm.
    2. A robot must obey the orders given it by human beings except where such orders would conflict with the First Law.
    3. A robot must protect its own existence as long as such protection does not conflict with the First or Second Laws.


    OK, if a robot learns that there are humans and there are other creatures that look and talk like humans but do not talk exactly like humans they are not human and can be killed. Imagine defining humans as some one that (in your best West Texas Redneck accent) "tawks lak dis" and that all others are non-human.
    OK, if a robot learns that humans ONLY LIVE in one place (in this case a specific planet) and that all space ships that come there contain only non-humans - they can be destroyed. Imagine defining humans as someone that ONLY DRIVES a white, 6 cylinder, with a bumper hitch Ford F-100 pickup and that all others are non-human.

    You control education and you control the world.

    Also in one of his post-Foundation novels, some robots came up with the Zeroth law:
    0. A robot may not harm humanity, or, by inaction, allow humanity to come to harm.

    [Begin My Opinion]
    So - AI?? for those of us who are pure of heart - it's something we could use and control, for all others, they'll use it to create a Big Brother world that ultimately would lead to a Terminator - Skynet world.

    I don't hold much hope. The pure of heart don't seem to do well in politics so we're screwed.
    [End My Opinion]
     
    Atlas, twp and hypnos like this.
  6. hypnos

    hypnos Moderator Staff Member

    Joined:
    Aug 7, 2018
    Messages:
    1,961
    Likes Received:
    2,370
    That may be true, but anyone including you can create your own AI... the commies don't monopolize tech.

    Great post.
     
    Atlas and twp like this.
  7. twp

    twp Administrator Staff Member

    Joined:
    Jun 16, 2018
    Messages:
    3,771
    Likes Received:
    4,288
    So we need a R. Daneel Olivaw to watch over us...
     
    WolfBrother, Atlas and hypnos like this.
  8. hypnos

    hypnos Moderator Staff Member

    Joined:
    Aug 7, 2018
    Messages:
    1,961
    Likes Received:
    2,370
    Do we need?

    I'll focus on that.

    At the pace that things are changing and evolving in our political environment and world in general, I see it sort of like as you said "interstellar space travel" you can try to do it manually without the aid of technology, but then why are you doing interstellar travel in the first place?

    If the destination is our survival, and the survival of our values, it would make sense to have the best tools at our disposal to help ensure it.

    I'm not saying we need to have it make our decisions. I am saying it may be helpful to have an accurate means of navigating through the asteroid field.

    I don't want an AI president. But I would like an AI watch dog.

    If for example we added a "rogue flow" extension to our browsers. Maybe its possible we could use it to aid in randomization of our online presence. Our data points, searches, habits, thought patterns are pretty much who we are in terms of our framework and personhood is concerned digitally. A AI that could help throw crawlers, google, other AI off the trail when it begins to see a pattern develop could be useful, or am I insane?

    I am not for unlimited application of A.I. I think that's a really terrible idea, we don't want machines pulling triggers and pushing certain buttons. But, I am solidly for using machines to extend the limit of human capability. Just like every tool has, as long as there have been tools.
     
    Atlas and twp like this.
  9. twp

    twp Administrator Staff Member

    Joined:
    Jun 16, 2018
    Messages:
    3,771
    Likes Received:
    4,288
    Given the demonstrated nature of humans, perhaps some non-human oversight might help?

    But where to place one's trust... Aye, there's the rub.
     
    Atlas and hypnos like this.
  10. hypnos

    hypnos Moderator Staff Member

    Joined:
    Aug 7, 2018
    Messages:
    1,961
    Likes Received:
    2,370
    I don't want "active" AI affecting its environment and making decisions , but I would like passive AI that gathers information and automates complicated and boring processes like research, and defeating captchas.
     
    Atlas and twp like this.
  11. hypnos

    hypnos Moderator Staff Member

    Joined:
    Aug 7, 2018
    Messages:
    1,961
    Likes Received:
    2,370
    I don't put any lasting trust in the physical world at all. But I like tools! :)

    These ads and interactive things to make decisions do more than generate ads, they capture snapshots of your personality profile among other things. I look at all these little YOUTUBE skip this ad, would you consider buying this product, questions like a MNPI, because they probably are.

    It would also be extremely fun to have an AI take paid surveys all day while I'm at work and make me some money...24 hours a day...
     
    Atlas likes this.
  12. hypnos

    hypnos Moderator Staff Member

    Joined:
    Aug 7, 2018
    Messages:
    1,961
    Likes Received:
    2,370
    @twp , yes a little non human guidance may be more than welcome as the stupidity rages on. I can see the appeal also in that respect
     
    Atlas likes this.
  13. hypnos

    hypnos Moderator Staff Member

    Joined:
    Aug 7, 2018
    Messages:
    1,961
    Likes Received:
    2,370
    https://gizmodo.com/if-you-solve-this-math-problem-you-could-steal-all-the-1836047131

    How handy would it be, to have a machine working this problem? Could help make you rich simply knowing the best time to buy and sell stocks or crypto based on historical market data, news headlines, and other indicators.

    Can I do it? (Not the p=np , that's not going to happen) Only as well as I am able to pay attention.

    A machine could help, allot.
     
    Atlas and twp like this.
  14. twp

    twp Administrator Staff Member

    Joined:
    Jun 16, 2018
    Messages:
    3,771
    Likes Received:
    4,288
    If it can defeat captchas, it gets my support.
     
    Atlas and hypnos like this.
  15. hypnos

    hypnos Moderator Staff Member

    Joined:
    Aug 7, 2018
    Messages:
    1,961
    Likes Received:
    2,370
    :D

    It would be nice! I hate those dumb things.
     
    Atlas and twp like this.
  16. WolfBrother

    WolfBrother Active Member

    Joined:
    Jul 7, 2019
    Messages:
    231
    Likes Received:
    542
    If it came down to it - yes.

    Spoiler Alert:
    I don’t know if you read the pre/post Foundation books but in the end Asimov ultimately merged the Robot series with the Foundation series.

    R. Daneel Olivaw ultimately merged his next generation positronic brain with a member of the planet who first defined all non-this planet spacecraft as absolutely not capable of being human (and therefore must be destroyed as an immediate danger to the planets humans) and then defined for that planets robots as non-human anyone who didn’t talk like the humans of that planet as non-human (and therefore must be destroyed as an immediate danger to the planets humans). Made me wonder a bit.
     
    Atlas, twp and hypnos like this.
  17. twp

    twp Administrator Staff Member

    Joined:
    Jun 16, 2018
    Messages:
    3,771
    Likes Received:
    4,288
    I don't remember reading those later Asimov stories. Got titles?
     
    hypnos and Atlas like this.
  18. WolfBrother

    WolfBrother Active Member

    Joined:
    Jul 7, 2019
    Messages:
    231
    Likes Received:
    542
    It’ll take a bit. I’ll have to look them up.
     
    hypnos likes this.
  19. WolfBrother

    WolfBrother Active Member

    Joined:
    Jul 7, 2019
    Messages:
    231
    Likes Received:
    542
    First place I looked.

    Titles in chronological order prequels, the original sequence, the sequels.

    Prelude to Foundation
    Forward the Foundation
    ——————————
    Foundation
    Foundation and Empire
    Second Foundation
    ——————————
    Foundations Edge
    Foundation and Earth

    Prequels - mainly about Hari Seldon and how he perfected his work.

    Sequels - ultimately about tying the Robot series to the Foundation series.

    Some people like only the original trilogy. Some prequel and not sequel. Some the other way around.
     
    hypnos and Atlas like this.
  20. twp

    twp Administrator Staff Member

    Joined:
    Jun 16, 2018
    Messages:
    3,771
    Likes Received:
    4,288
    Ok, thanks, I've not read the Prequels or Sequels books.
     
    hypnos and Atlas like this.

Share This Page

LINK CODE
We post links, such as those above, to websites which we like and which have content compatible with our website.

We particularly appreciate having a reciprocal link agreement with you, in which you agree to place a link back to us on your website.
For that purpose, we offer this HTML link code which you may place in your website.
Note that our logo image size defaults to 300 pixels width and 50 pixels height.
It looks like this:

You may use copy/paste in this box.