Is AI going to change the world?

by Reasonfirst 54 Replies latest jw friends

  • Reasonfirst
    Reasonfirst

    Another development:

    AI system self-organizes to develop features of brains of complex organisms

    Date:
    November 20, 2023
    Source:
    University of Cambridge
    Summary:
    Scientists have shown that placing physical constraints on an artificially-intelligent system -- in much the same way that the human brain has to develop and operate within physical and biological constraints -- allows it to develop features of the brains of complex organisms in order to solve tasks
    https://www.sciencedaily.com/releases/2023/11/231120124246.htm


    Quote:

    With their system, however, the physical constraint meant that the further away two nodes were, the more difficult it was to build a connection between the two nodes in response to the feedback. In the human brain, connections that span a large physical distance are expensive to form and maintain.

    When the system was asked to perform the task under these constraints, it used some of the same tricks used by real human brains to solve the task. For example, to get around the constraints, the artificial systems started to develop hubs -- highly connected nodes that act as conduits for passing information across the network.

    More surprising, however, was that the response profiles of individual nodes themselves began to change: in other words, rather than having a system where each node codes for one particular property of the maze task, like the goal location or the next choice, nodes developed a flexible coding scheme.



  • Reasonfirst
    Reasonfirst

    Maybe this will become a problematical area for AI.

    https://asiatimes.com/2023/11/ai-plus-robotics-equals-fierce-new-weaponry/

    Imagine armed AI+ robots cruising the streets as police auxiliaries?

    Imagine an AI+robot in the hands of crime gangs?

  • TonusOH
    TonusOH

    Neural networks have been used to mimick brain activity for decades. I remember reading an article about it in the 90s. Interesting stuff, but not quite the same thing in my mind.

  • Reasonfirst
  • enoughisenough
    enoughisenough

    I don't remember the movie, but when I think of this A1 business, I think of "dissasemble #5" So many years since I saw the movie, but the robots decided they did not want to be dissasembled and sought to protect themselves. There seems to be some merit to the thought of Hollywood and future developements. ( that is scary since there are movies about sombies AND the CDC considers that can become a reality.)

  • Reasonfirst
    Reasonfirst

    There's an very interesting court case coming up in the USA. The NY TImes is suing Microsoft and OpenAI, ( the creator of ChatGPT) claiming misuse of articles published by the NY Times.

    It brings into the open, an interesting aspect of AI, as the NYTimes case points out, AI has to be fed data - which I guess is exactly what happens with humans through various forms of education and training, including, of course, religious ideas.

    I have little understanding of how Microsoft and OpenAI may have "misused" the NY Times copyrighted information, but perhaps it means that data fed to an AI program can be slanted - again just as in the example of the human mind.

    A human mind fed with biased information can refuse to examine contrary ideas, our experiences as JW's (or, any other religion, for that matter) illustrates the problem, Rubbish in - rubbish out, some call it.

    Will AI be like that???

    FYI, I reads about this case, on an Australian Media service, the ABC, a government owned (but, supposedly independent) media group.

    Check it at: https://www.abc.net.au/news/2023-12-28/new-york-times-sued-microsoft-bing-chatgpt-openai-chatbots/103269036


  • SydBarrett
    SydBarrett


    "In the near future, only jobs that can be distilled into algorithims are likely to be replaced."

    The problem is that a significant percentage of the population earns their living in such jobs. Yes, jobs have always been replaced by technology. What's worrisome is the speed at which it may happen. Anyone who has ever worked in a large corporate office environment knows that probably half of the jobs are monotonous "drone" jobs. But the low level stuff still provides health insurance, a lower middle class income and a 401k.

    I dont think our current political system is prepared or even willing to prepare to deal with 25% unemployment suddenly happening. It may be a short term anamoly as far as history goes, but that level of instability can be very dangerous.

    Maybe they will all go into real estate and play musical chairs flipping houses to one another.


  • Anony Mous
    Anony Mous

    @SydBarrett: the automatic loom and the steam engine did pretty much the same thing, probably at greater pace than so-called AI can. Unemployment became lower and overall wealth and quality of life higher as a result and today there are now millions of open jobs to do maintenance on the machines. We also have shortages of people that can build an AI-oriented infrastructure, both physically and logically, the explosion in chip manufacturing that will be required, it will be more like the Internet when it became well established, everyone became a web developer overnight.

    In their current state, the current models may be good enough to augment a good entry level employee in coding, accounting etc, it’s a better search engine and may be able to replace ghostwriters and anyone producing such derivative crap. So white collar jobs are more likely to be replaced, AI plumbers, electricians etc will require AGI and very advanced robots which humanity is at least a century away from if at all possible to do economically.

  • slimboyfat
    slimboyfat

    If you can get past the terrible screechy voice I think this debate was excellent. The woman made many great points and the guy was utterly clueless. He got completely sidetracked by whether AI can be conscious (it can’t) and whether it will ever work the same way as a human brain (it won’t). It doesn’t need to be conscious or like a human brain in order to be dangerous, quite the reverse. If anything I think the woman wasn’t strident enough in laying out the dangers. Our situation is probably significantly worse than even she was arguing. Maybe she didn’t want to be too pessimistic for the audience.

    https://www.youtube.com/live/1rnam1w8ztM?si=vMcvnYTU-_J81t5t

  • Vidiot
    Vidiot

    Meh.

    As long as some dumbass doesn’t name it “Skynet”, we should be fine.

Share this

Google+
Pinterest
Reddit