Bill Gates is worried about artificial intelligence too

Discussion in 'The Thunderdome' started by VolDad, Jan 29, 2015.

  1. VolDad

    VolDad Super Moderator

    Bill Gates has a warning for humanity: Beware of artificial intelligence in the coming decades, before it's too late.

    Microsoft's co-founder joins a list of science and industry notables, including famed physicist Stephen Hawking and Internet innovator Elon Musk, in calling out the potential threat from machines that can think for themselves. Gates shared his thoughts on AI on Wednesday in a Reddit "AskMeAnything" thread, a Q&A session conducted live on the social news site that has also featured President Barack Obama and World Wide Web founder Tim Berners-Lee.

    "I am in the camp that is concerned about super intelligence," Gates said in response to a question about the existential threat posed by AI. "First, the machines will do a lot of jobs for us and not be super intelligent. That should be positive if we manage it well. A few decades after that, though, the intelligence is strong enough to be a concern."


    http://www.cnet.com/uk/news/bill-gates-is-worried-about-artificial-intelligence-too/
     
  2. NorrisAlan

    NorrisAlan Super Moderator

  3. NorrisAlan

    NorrisAlan Super Moderator

    I do not fear artificial intelligence as much as I fear human de-evolution if we simply start letting machinesdo everything for us.
     
  4. droski

    droski Super Moderator

    I approve of skynet
     
  5. IP

    IP Advanced Pruitt Apologetics Bot

    An artificial intelligence with no dependence is pretty much game over for us.
     
  6. NorrisAlan

    NorrisAlan Super Moderator

    I am just not certain we will ever be able to create a mechanical intelligence.
     
  7. IP

    IP Advanced Pruitt Apologetics Bot

    I think that's ego. I don't know how my computer works exactly, but I built it myself. So it could be with intelligence and consciousness. We already know it is an emergent property in terms of the human brain and can't be broken down into units the way the architecture of the physical brain can be. Similarly, we may not know exactly when we produce true AI until after the fact.
     
  8. gcbvol

    gcbvol Fabulous Moderator

    May not happen as soon as some expect (or at all), but when guys like Hawking, Musk and Gates openly discuss their concerns it certainly gives the probability merit.
     
  9. utvol0427

    utvol0427 Chieftain

    I welcome our artificially intelligent overlords.
     
  10. IP

    IP Advanced Pruitt Apologetics Bot

    I really think people give intelligence too much credit, if they think it could not be produced artificially. The issue is that an AI wouldn't be constrained by the slow rate of biological evolution, and could exponentially improve itself rapidly.
     
  11. IP

    IP Advanced Pruitt Apologetics Bot

    Why would they be our overlords? We'd literally be parasites or vermin. They'd wipe us out.
     
  12. justingroves

    justingroves supermod

    Ultron
     
  13. NorrisAlan

    NorrisAlan Super Moderator

    I don't think that they would immediately come to the conclusion "wipe them out" unless it felt threatened (a la Skynet that nuked the planet to keep itself from being unplugged). Especially if emotion is an inevitable by-product of intelligence (which it might not be).

    That said, a mechanical brain might just be too hard to produce using modern hardware and theory.
     
  14. Volst53

    Volst53 Super Moderator

    Wouldn't emotion be a weakness.
     
  15. NorrisAlan

    NorrisAlan Super Moderator

    How so?

    But that said, humans have sociopaths and other social disorders, so perhaps it is possible to have consciousness without emotions. And of course, I am coming at this with a human bias as well. As IP said, it is my ego demanding that all intelligence and consciousness be like my own. Perhaps a Data is possible.
     
  16. IP

    IP Advanced Pruitt Apologetics Bot

    Well, we'd be either competition for resources or a direct nuisance/existential threat. So... I think it would rather quickly go Skynet.
     
  17. IP

    IP Advanced Pruitt Apologetics Bot

    I think a Data is as equally possible as a Lore or Skynet. Not a gamble worth taking.
     
  18. JohnnyQuickkick

    JohnnyQuickkick Great Man

    well if Microsoft makes it it'll probably crash a lot anyway
     
  19. IP

    IP Advanced Pruitt Apologetics Bot

    You're not getting what an AI is. You make it... and then it can remake itself. Over and over and over again, improving, revising, expanding itself faster than humans ever could. The first AI will be harmless. But the AI's it produces, and it's AI's produce... And that could happen all in less than a day. Or an hour. Or maybe a minute.
     
  20. JohnnyQuickkick

    JohnnyQuickkick Great Man

    It was a joke, Sheldon
     

Share This Page