11 votes

How the Enlightenment ends: Philosophically, intellectually—in every way—human society is unprepared for the rise of artificial intelligence

Posted by Deimos
Tags: artificial intelligence, society, enlightenment, humans, author.henry a kissinger, source.the atlantic, paywall.soft
https://www.theatlantic.com/magazine/archive/2018/06/henry-kissinger-ai-could-mean-the-end-of-human-history/559124/

Link information

This data is scraped automatically and may be incorrect.

Title
How the Enlightenment Ends
Authors
Henry A. Kissinger
Published
May 15 2018
Word count
2628 words

7 comments

  1. [2]
    Flashynuff
    Link
    Oh wow, I didn't even realize until the end of the article that this was written by Henry Kissinger. Wild. This was a really good read, and I think it goes over a lot of the important questions AI...

    Oh wow, I didn't even realize until the end of the article that this was written by Henry Kissinger. Wild.

    This was a really good read, and I think it goes over a lot of the important questions AI engineers should be asking. We always need to be looking at the ethical implications of new tech and I think AI far surpasses any previous tech in the amount of ethical conundrums it raises. How we hold an AI that makes decisions based on millions of iterations accountable for those decisions if we decide they're wrong?

    A short story I always like to recommend when it comes to AI is Metamorphosis of the Prime Intellect, by Roger Williams. (Fair warning: the first chapter veers into a graphic sex scene. It makes sense in the context of the story, later, but is a little shocking). It deals with the issue of what happens when an AI is so intelligent, it figures out how to bend the rules of the universe.

    4 votes
    1. SourceContribute
      Link Parent
      I saw the byline and avoided reading this article at first...but now I'm reconsidering, will give it a proper read because it does seem to reflect on history.

      I saw the byline and avoided reading this article at first...but now I'm reconsidering, will give it a proper read because it does seem to reflect on history.

      3 votes
  2. [4]
    BuckeyeSundae
    Link
    I end up being pretty skeptical about these sorts of arguments, which people have been whinging about for about as long as Henry Kissinger has been an important voice. This technology will only...

    I end up being pretty skeptical about these sorts of arguments, which people have been whinging about for about as long as Henry Kissinger has been an important voice. This technology will only take root where it's computationally effective to allow it to have the reigns. We will try to press the edge of what's possible to prove we can do it, but as far as the power granted to these programs? It's unlikely anyone will give a single program the power over a country's nuclear arsenal or something.

    When it comes to areas where value judgments are required, it's still very unlikely that we will agree as a society (in most any country) to grant any AI power to make those judgments instead of us. I mean, necessary qualification of 2016, Trump-did-it, etc., but I don't think anyone trusts the people programming anything to be so thorough as to allow for neutral, value-based decision making in an AI program that has the power to decide when to terminate a national policy or something similar.

    Healthcare is different. I'm a bit worried about how AI is being used in healthcare, because that does get into the sort of life-or-death decision making about where to distribute resources. But it's still not the sort of society-changing-doomsday scenario that we're fretting about in our highest Sci-fi stories.

    3 votes
    1. [2]
      SourceContribute
      Link Parent
      Asimov wrote a story about this, "All the Troubles of the World" where a computer manages the world's economy

      Asimov wrote a story about this, "All the Troubles of the World" where a computer manages the world's economy

      3 votes
      1. BuckeyeSundae
        Link Parent
        Oh sure, I see plots where an AI takes is permitted to manage human affairs all the time. "I Have No Mouth and I Must Scream" comes immediately to mind where the AI is in charge of resolving all...

        Oh sure, I see plots where an AI takes is permitted to manage human affairs all the time. "I Have No Mouth and I Must Scream" comes immediately to mind where the AI is in charge of resolving all human conflict and resolves the most efficient way to do that is by killing all humans but for personal (rage) reasons, the AI saves five humans who played a role in his existence and keeps torturing them for the foreseeable future.

        As far as the real world is concerned though, I'm not convinced we are close to a meaningful consciousness in digital form, let alone anything that could make value judgments for humans.

        3 votes
    2. cfabbro
      Link Parent
      I am skeptical too. And there definitely is a lot of baseless supposition, generational denigration and "old man yells at clouds" in certain parts of the article. E.g. I also find it kind of...

      I am skeptical too. And there definitely is a lot of baseless supposition, generational denigration and "old man yells at clouds" in certain parts of the article. E.g.

      All of these pressures weaken the fortitude required to develop and sustain convictions that can be implemented only by traveling a lonely road, which is the essence of creativity.

      Users of the internet emphasize retrieving and manipulating information over contextualizing or conceptualizing its meaning. They rarely interrogate history or philosophy; as a rule, they demand information relevant to their immediate practical needs.

      I also find it kind of ironic that Kissinger is worried about ethics and value judgment in AI development when he himself is an amoral, cruel, heartless, political animal... But he isn't stupid and this article does make lots of good points as well. Specifically about how value judgement is derived in AI currently and how counter productive and potentially dangerous the current trend of black box "we don't understand why it's making these decisions" neural nets is.

      1 vote