15 votes

What will "emergent" tech make our lives look like in a few years?

I've been thinking about this a lot lately, and I am honestly just curious as to what Tildes thinks.

With advances in generative AI, which will likely only get better, along with the push for agents, what would the internet look like? Something that I grew up with basically my entire life, and I attribute to a lot of great socialization and exposing me to new ideas, might die or become completely unrecognizable. It was already becoming bad before generative AI launched to the public, with bots and shill accounts everywhere. Now, many times, I have no idea how I will know if I will be speaking to an actual person. Is this something we can even solve, at least without massively breaching people's privacy?

@boxer_dogs_dance posted this article about human certification for books. Are we going to need something for necessitating and verifying human engagement over the internet?

I consider myself a critical optimist about the future overall. I think Deep Learning and Neural Network developments may assist with massive healthcare breakthroughs, helping us find medicines and cures that may have taken an incredibly long time. However, I also feel that it's going to be an incredibly rocky road on the way there.

Other things as well. Robotics and VR/AR/XR. How are these going to change our lives? I think change will be slow, but will how we engage with technology completely change? While current AR offerings are limited, they will likely continue to improve. Imagine a truly useful AR device as something that would feel as natural as a pair of everyday sunglasses, capable of seamlessly overlaying information without the bulk or battery limitations of today’s prototypes. It could give you information about the world around you in real time, hell just imagine it translating things around you.

Will I have an Uber driver in a decade, or will it be fully automated? Will we continue to become more isolated in terms of physical contact, but then retreat into digital worlds?

What should the next generation even learn at this point? Education is always slow and last to change. If you had a child now, how would you prepare them for a future? How should schools be changing? Will they even be able to keep up with sweeping waves? There are financial realities at play here, what jobs may even be left for someone to thrive in, not just scrape by?

Consider this post just general musings, and really just wanting to hear what other people think about this.

11 comments

  1. hobblyhoy
    Link
    I think the concept of being anonymous while online will die. Some kind of human authentication system will become standard on every major virtual space. There will be many places which hold out...

    I think the concept of being anonymous while online will die. Some kind of human authentication system will become standard on every major virtual space. There will be many places which hold out against this but the AIs/agents built to crawl, register themselves, and begin pushing subtle spam and politically influential comments will make the signal to noise too poor for all but the oblivious to continue using.

    I also suspect that the human authentication system will condense to a single company or few major players which you register once and that can be applied to any other number of sites, likely as an extension of the existing single sign on providers.

    Both of these would suck and kill off a large part of the Internet for me.

    18 votes
  2. [5]
    creesch
    (edited )
    Link
    I have been interested in tech and following development of it closely since my teenage years. Meaning that I have been following it for over 25 years now. The only thing I can say with certainty...

    I have been interested in tech and following development of it closely since my teenage years. Meaning that I have been following it for over 25 years now. The only thing I can say with certainty is that we have no clue what the future will bring. I have seen seemingly promising technology never go mainstream while things that did seem gimmicky at first become widely adopted.

    As far as your concern goes about human interaction. I am sorry to burst your bubble, but it already has been bad for well over a decade. AI has made things ever so slightly more bad, but we have had content farms dedicated to SEO for a long time already. The same goes for online interactions. Moderating reddit for a decade also has shown me how many bad faith actors there are and while we now have “bots” before we had hordes of cheap labor in India doing pretty much the same thing.

    I think a lot of people were less aware as it was a bit more subtle. But I already started to tailor my Google queries and general approach to information finding years before AI really entered the picture. I suppose that is also why I don't think Kagi is that amazing when I tried it. It removes some of the more obvious bad results, but there is a still of the same bullshit I have been seeing for years now. So I might as well keep using Google and effectively get the same results.

    However, if we go with human certification for engagement over the internet. What does count as that? With LLMs becoming more mainstream and some people heavily leaning on them to augment their writing does that still count as human engagement?

    As far as VR/AR/XR goes, we'll see. These are technologies that have been on the horizon since at least the 90s and always seem slightly out of reach for mainstream adaption. I do think we are at a point where the technology is (close to) good enough and where price is one of the biggest barriers. But adoption also really depends on other factors like ease of use, vendor lock in, etc. There have been fairly good products in these categories that have been dead on arrival as they only worked in a very limited eco system.

    Education is always slow and last to change.

    Eh, that is also a misunderstanding of education or maybe the US having a failing education system (I assume you are from the US anyway). Good education while still educating you on subjects, primarily should focus on the underlying skills and general skills that help you further develop what you learn in school/college. Nobody leaves college an expert, they leave with a basis to start working at the bottom of their career path and further develop the skills and expertise.
    In fact, I do think that calls for education to change to better fit "highly specific subject matter" are damaging for an education system as they tend to focus on rote learning. Mind you, I am not saying that education systems shouldn't change. But change should be aimed at the basics of teaching students to be capable adaptive human beings. The subject is part of that but only as a basis, the primary focus should be on the skills to further build on that basis outside of the education system.

    10 votes
    1. vord
      Link Parent
      Also, it completely ignores the absolutely critical role that universities play in doing the hard sciences. Almost every single major technological advancement of the past hundred years has its...

      Also, it completely ignores the absolutely critical role that universities play in doing the hard sciences. Almost every single major technological advancement of the past hundred years has its roots in research universities (and government sponsored institutions which are fed from grad programs).

      The mRNA tech used in the COVID vaccines? Invented at University of Pennsylvania, not from the pockets of Moderna and Pfizer.

      5 votes
    2. [3]
      ShinRamyun
      Link Parent
      Thanks for sharing your thoughts. How have you been tailoring your Google queries and general approach to information finding? Honestly, I switched to Kagi and I found it has been a lot better,...

      Thanks for sharing your thoughts.

      How have you been tailoring your Google queries and general approach to information finding? Honestly, I switched to Kagi and I found it has been a lot better, though there Assistant offering is fairly mediocore.

      1 vote
      1. [2]
        creesch
        Link Parent
        It's hard to describe because I have been doing it for so long. But I'll give it a shot. I am going to focus on technical research for things like software development, Linux, etc. Simply because...

        It's hard to describe because I have been doing it for so long. But I'll give it a shot. I am going to focus on technical research for things like software development, Linux, etc. Simply because for local searches Google is still much better than Kagi so for that you don't need me going into detail. There are also some other exceptions I will go into further down*.

        For starters, a search engine for me is not the primary means of searching for information. It simply is a means to get to information if I am not quite sure where to get it. So, I often start out with official documentation, GitHub, etc. Which will get you pretty far in a lot of cases and I feel like is a blind spot for a lot of people.
        Github issues for example can be really handy to see if people have had the same issue. You just need to know to also search for closed issues.

        If I am not entirely sure what resources are available on a given subject I might try a broader search and see what pops up. I do have a good sense of what is SEO spam and what is might be genuine resources.

        Once I am more aware of the resources available, I might also include the resources specifically in my search query. For example if it is web development related I might append MDN to the search term. Certainly when my search is still broad I tend to do this to see if I can find more specific terminology to further refine my search.

        If I am using Google for something specific I will make sure I am actually searching for that specific phrase and not Google's interpretation. Just yesterday I was dealing with an odd issue at work. Effectively a third party web application hosted in our environment was working almost perfectly with some buttons being grayed out. Now, in this case, when you try to search for Application name button description not available you will get no useful results. However, if you search for Application name "exact text of button" not available you suddenly will see github issues pop up about this very specific issue. The reason for that is because enclosing things in double quotation marks makes google search for that exact phrase. Not only that, it makes a required phrase to be part of the search results.

        It does require you to have a specific text to search for. But when dealing with software that is often the case. You are either dealing with a specific error, specific function, etc. So, making your search more specific automatically excludes a ton of general SEO spam as they are aimed at more general less specific search queries.

        And to be clear, even in the latter case I don't expect Google or any search engine to directly lead me to an answer. I only expect it to lead me to actual information which might be an answer or enough of a lead to find the answer in other resources. While I am writing this out I am also reminded about this comment I wrote yesterday about the use of LLMs. Which makes me wonder if this is a similar case of people depending on search engines on a much deeper level than I ever did. To the point where the steps of doing research without a search engine presenting everything right away is a skill some people never learned or maybe lost?
        To be clear, I am not saying that things haven't gotten worse over the years. They certainly have, but like I said for how I use search engines Kagi doesn't seem to actually provide much in the way of improvements over google.

        *Some other exceptions I should note that some of my habits don't transfer as easily because they lean on various other sources not available in other countries. For example, in the Netherlands we have a tech website called [Tweakers, which has a pricewatch feature](https://tweakers.net/pricewatch/) which is sort of like newegg but then only listing items from other stores. It effectively contains all webshops in the Netherlands that deal in tech. If a shop is not on there then it might as well not exist. While not perfect (there are sponsored reviews) for the most part it is quite okay, reliable and has very decent search and filtering options.
        6 votes
        1. arghdos
          Link Parent
          I don’t think this is strictly true anymore, to get this behavior you have to flip on Tools>Verbatim in the search results… which sucks. It does still seem to hint the phrase which is often good...

          double quotation marks makes google search for that exact phrase. Not only that, it makes a required phrase to be part of the search results.

          I don’t think this is strictly true anymore, to get this behavior you have to flip on Tools>Verbatim in the search results… which sucks. It does still seem to hint the phrase which is often good enough tho

          2 votes
  3. [2]
    PuddleOfKittens
    Link
    It really depends on the limits on our imagination - not in the sense that "we're almost unlimited!", more that we set stupid goals and then are surprised that they're expensive to achieve. I...

    It really depends on the limits on our imagination - not in the sense that "we're almost unlimited!", more that we set stupid goals and then are surprised that they're expensive to achieve.

    I think there's a really interesting effect where computers (including sigh AI) are making our tools increasingly capable, effectively reducing the minimum size of "a factory" down to a handful of people, or even just one person.

    That might sound trivial, but I want to introduce two concepts: the minimum size for a self-sufficient industrial economy, and BATNA (Best Alternative To Negotiated Agreement).

    Basically, if you can create a (mostly) self-sufficient industrial economy with e.g. 100 people, then you can choose to just... not participate in the economy. There's a difference between this stuff being more expensive versus impossible. I added the "mostly" to the start of this paragraph because some stuff, like computer parts, might never be particularly effective to produce in a decentralized fashion. But, if your machines all need an e.g. $10 chip each, then for $1000 you can buy quite a long time's worth of chips for basically nothing! (ok realistically maybe $10k or $100k, but you get the point).

    Okay, so what does that mean?

    Well, it means that without explicit coercion (i.e. roll in the tanks), everyone will have a better BATNA than ever before. And game-theoretically, BATNA is the single biggest determinant of bargaining power in any negotiation. Which means, theoretically, society will have to provide a bigger carrot to more local governments (at the expense of federal govts) to prevent people from just leaving. Assuming they don't roll in the tanks.

    6 votes
    1. vord
      Link Parent
      It's a neat idea, but I've seen enough into the manufacturing space that I don't see that inside of 50 years. It might only take 1 person to run the factory, but it can take thousands of people to...

      It's a neat idea, but I've seen enough into the manufacturing space that I don't see that inside of 50 years. It might only take 1 person to run the factory, but it can take thousands of people to design and build one in the first place.

      Machines which modify physical matter are exponentially more difficult to keep working. Hence why printers all suck.

      That, and while efficiencies of scale allow 2% of the population to feed our industrial society, that ratio does not hold down to the hundreds...2 or 3 people can't grow enough food for 100. We'd probably need on the order of 15.

      4 votes
  4. [3]
    nic
    Link
    The internet put the wealth of all human knowledge at our fingertips, it gave knowledge workers the ability to work more productively and from home. Most of the sci fi books assumed this would...

    The internet put the wealth of all human knowledge at our fingertips, it gave knowledge workers the ability to work more productively and from home. Most of the sci fi books assumed this would make us smarter and happier. Instead it feels to me like it made us stupider, angrier, unhappier.

    AI/VR in theory should make us smarter and happier. But I fear it will make us more isolated, more likely to dismiss facts in favor of fake news that supports our preferred world view, and overall increase wealth inequality until we have a violent uprising or some sort of catastrophic world war.

    Uber drivers are definitely going away.

    2 votes
    1. [2]
      Flapmeat
      (edited )
      Link Parent
      The Internet gave us everything we wanted at our fingertips and we mostly used it to serve our base instincts. It's kinda gross. We use it to reassure our egos, instead of learning about others....

      The Internet gave us everything we wanted at our fingertips and we mostly used it to serve our base instincts.
      It's kinda gross.

      We use it to reassure our egos, instead of learning about others.
      We use it to "jerk off" instead of becoming a part of irl communities.
      Maybe that's a value judgement but like, is the Pinnacle of human achievement people blankly scrolling dimly lit "content treadmills" as we mainline salty, sweet nutrition sludge into the back of our heads ?

      2 votes
      1. nic
        Link Parent
        If it makes you feel better, Youtube's algorithms are feeding my kid with highly educational videos that taught him how to cook, to the point where he was a more advanced cook at 12 than either of...

        If it makes you feel better, Youtube's algorithms are feeding my kid with highly educational videos that taught him how to cook, to the point where he was a more advanced cook at 12 than either of his parents.

        All we had to do was limit his time on the gaming treadmill, and luckily he naturally prefers intellectually stimulating content treadmills. I know there must be a lot of people like him, but there don't seem to be many in his classes.

        4 votes