8 ms·

What fascinates me about your comment is that you are expressing that you trusted what you read before. For me, LLMs don't change anything. I already questioned

• baq 2y ago
  scale makes all the difference. society without trust falls apart. it's good if some people doubt some things, but if everyone necessarily must doubt everything, it's anarchy.
  • dangitman 2y ago
    [dead]
    • Hoasi 2y ago
      > Is our society built on trust? A good part of society, the foundational part, is trust. Trust between individuals, but also trust in the sense that we expect things to behave in a certain way. We trust things like currencies despite their flaws. Our world is too complex to reinvent the wheel whenever we need to do a transaction. We must believe enough in a make-believe system to avoid perpetual collapse.
      • dangitman 2y ago
        [dead]
  • vouaobrasil 2y ago
    Perhaps that anarchy is the exact thing we need to convince everyone to revolt against big tech firms like Google and OpenAI and take them down by mob rule.
• kombookcha 2y ago
  Debunking bullshit inherently takes more effort than generating bullshit, so the human factor is normally your big force multiplier. Does this person seem trustworthy? What else have they done, who have they worked with, what hidden motivations or biases might they have, are their vibes /off/ to your acute social monkey senses? However with AI anyone can generate absurd torrential flows of bullshit at a rate where, with your finite human time and energy, the only winning move is to reject out of hand any piece of media that you can sniff out as AI. It's a solution that's imperfect, but workable, when you're swimming through a sea of slop.
  • bitexploder 2y ago
    Maybe the debunking AIs can match the bullshit generating AIs, and we will have balance in the force. Everyone is focused on the generative AIs, it seems.
    • nicce 2y ago
      There is always more money available for bullshit generation than bullshit removal.
    • desdenova 2y ago
      No, they can't. They'll still be randomly deciding if something is fake or not, so they'll only have a probability of being correct, like all nondeterministic AI.
  • ontouchstart 2y ago
    Debugging is harder than writing code. Once the code passed linter, compiler and test, the bugs might be more subtly logical and require more effort and intelligence. We are all becoming QA of this super automated world.
• nicce 2y ago
  In the past, you had to put a lot of effort to produce a text which seemed to be high quality, especially when you knew nothing about the subject. By the look of text and the usage of the words, you could tell how professional the writer was and you had some confidence that the writer knew something about the subject. Now, that is completely removed. There is no easy filter anymore. While the professional looking text could have been already wrong, the likelihood was smaller, since you usually needed to know something at least in order to write convincing text.
  • roenxi 2y ago
    > While the professional looking text could have been already wrong, the likelihood was smaller... I don't criticise you for it, because that strategy is both rational and popular. But you never checked the accuracy of your information before so you have no way of telling if it has gotten more or less accurate with the advent of AI. You were testing for whether someone of high social intelligence wanted you to believe what they said rather than if what they said was true.
    • dietr1ch 2y ago
      I guess the complaint is about losing this proxy to gain some assurance for little cost. We humans are great at figuring out the least amount of work that's good enough. Now we'll need to be fully diligent, which means more work, and also there'll be way more things to review.
      • roenxi 2y ago
        I'd argue people clearly don't care about the truth at all - they care about being part of a group and that is where it ends. It shows up in things like critical thinking being a difficult skill acquired slowly vs social proof which humans just do by reflex. Makes a lot of sense, if there are 10 of us and 1 of you it doesn't matter how smartypants you may be when the mob forms. AI does indeed threaten people's ability to identify whether they are reading work by a high status human and what the group consensus is - and that is a real problem for most people. But it has no bearing on how correct information was in the past vs will be in the future. Groups are smart but they get a lot of stuff wrong in strategic ways (it is almost a truism that no group ever identifies itself or its pursuit of its own interests as the problem).
• tempfile 2y ago
  > I already questioned the information before and continue to do so. You might question new information, but you certainly do not actually verify it. So all you can hope to do is sense-checking - if something doesn't sound plausible, you assume it isn't true. This depends on having two things: having trustworthy sources at all, and being able to relatively easily distinguish between junk info and real thorough research. AI is a very easy way for previously-trustworthy sources to sneak in utter disinformation without necessarily changing tone much. That makes it much easier for the info to sneak past your senses than previously.
• thesz 2y ago
  Propaganda works by repeating the same in different forms. Now it is easier to have different forms of the same, hence, more propaganda. Also, it is much easier to iinfluence whatever people write by influencing the tool they use to write. Imagine that AI tools sway generated sentences to be slightly close, in summarisation space, to the phrase "eat dirt" or anything. What would happen?
  • ImHereToVote 2y ago
    Hopefully people will exercise more judgement now that every Tom, Dick, and Harry scam artists can output elaborate prose.
• eesmith 2y ago
  The negation of 'I cannot trust' is not 'I could always trust' but rather 'I could sometimes trust'. Nor is trust meant to mean something is absolute and unquestionable. I may trust someone, but with enough evidence I can withdraw trust.
• ffsm8 2y ago
  Trust as no bearing on what they said. Reading was a form of connecting with someone. Their opinions are bound to be flawed, everyone's are - but they're still the thoughts and words of a person. This is no longer the case. Thus, the human factor is gone and this reduces the experience to some of us, me included.
  • farleykr 2y ago
    This is exactly what’s at stake. I heard an artist say one time that he’d rather listen to Bob Dylan miss a note than listen to a song that had all the imperfections engineered out of it.
    • herval 2y ago
      The flipside of that is the most popular artists of all time (eg Taylor Swift) do autotune to perfection, and yet more and more people love them
      • kombookcha 2y ago
        If you ask a Swiftie what they love about Taylor Swift, I guarantee they will not say "the autotune is flawless". They're not connecting with the relative correctness of each note, but feeling a human, creative connection with an artist expressing herself.
        • herval 2y ago
          They're "creatively connecting" to an autotuned version of a human, not to a "flawed Bob Dylan"
          • kombookcha 2y ago
            They're not connecting to the autotune, but to the artist. People have a lot of opinions about Taylor Swift's music but "not being personal enough" is definitely not a common one. If you wanna advocate for unplugged music being more gratifying, I don't disagree, but acting like the autotune is what people are getting out of Taylor Swift songs is goofy.
• croes 2y ago
  The quota changed because it's now easier and faster
• tuyguntn 2y ago
  > For me, LLMs don't change anything. I already questioned the information before and continue to do so. I also did, but LLM increased the volume of content, which forces my brain first try to identify if content is generated by LLMs, which is consuming a lot of energy and makes brain even less focused, because now it's primary goal is skimming quickly to identify, instead of absorbing first and then analyzing info
  • desdenova 2y ago
    The web being polluted only makes me ignore more of it. You already know some of the more trustworthy sources of information, you don't need to read a random blog which will require a lot more effort to verify. Even here on hackernews, I ignore like 90% of the spam people post. A lot of posts here are extremely low effort blogs adding zero value to anything, and I don't even want to think whether someone wasted their own time writing that or used some LLM, it's worthless in both cases.
• everdrive 2y ago
  How do you like questioning much more of it, much more frequently, from many more sources? And mistrusting it in new ways. AI and regular people are not wrong in the same ways, nor for the same reasons, and now you must track this too, increasingly.
• voidmain0001 2y ago
  I read the original comment not as a lament of not being able to trust the content, rather, they are lamenting the fact that AI/LLM generated content has no more thought or effort put into it than a cheap microwave dinner purchased from Walmart. Yes, it fills the gut with calories but it lacks taste. On second thought, perhaps AI/LLM generated content is better illustrated with it being like eating the regurgitated sludge called cud. Nothing new, but it fills the gut.
• a99c43f2d565504 2y ago
  Perhaps "trust" was a bit misplaced here, but I think we can all agree on the idea: Before LLMs, there was intelligence behind text, and now there's not. The I in LLM stands for intelligence, as written in one blog. Maybe the text never was true, but at least it made sense given some agenda. And like pointed out by others, the usual text style and vocabulary signs that could have been used to identify expertise or agenda are gone.
  • danielmarkbruce 2y ago
    Those signs are largely bs. It's a textual version of charisma.
  • lmm 2y ago
    > Perhaps "trust" was a bit misplaced here, but I think we can all agree on the idea: Before LLMs, there was intelligence behind text, and now there's not. The I in LLM stands for intelligence, as written in one blog. Maybe the text never was true, but at least it made sense given some agenda. Nope. A lot of people just wrote stuff. There were always plenty of word salad blogs (and arguably entire philosophy journals) out there.
• rsynnott 2y ago
  There are topics on which you should be somewhat suspicious of anything you read, but also many topics where it is simply improbable that anyone would spend time maliciously coming up with a lie. However, they may well have spicy autocomplete imagine something for them. An example from a few days ago: https://news.ycombinator.com/item?id=41645282 https://news.ycombinator.com/item?id=41645282
• desdenova 2y ago
  Exactly. The web before LLMs was mostly low effort SEO spam written by low-wage people in marketing agencies. Now it's mostly zero effort LLM-generated SEO spam, and the low-wage workers lost their jobs.
  • vouaobrasil 2y ago
    The difference is that now we'll have even more zero-effort SEO spam because AI is a force multiplier for that. Much more.
• galactus 2y ago
  I think it is a totally different threat. Excluding adversarial behavior, humans usually produce information with a quality level that is homogeneous (from homogeneously sloppy to homogeneously rigurous). AI otoh can produce texts that are quite accurate globally with some totally random hallucinations here and there. It makes it quite harder to identify
• akudha 2y ago
  There were news reports that Russia spent less than a million dollars on a massive propaganda campaign targeting U.S elections and the American population in general. Do you think it would be possible before internet, before AI? Bad actors, poorly written/sourced information, sensationalism etc have always existed. It is nothing new. What is new is the scale, speed and cost of making and spreading poor quality stuff now. All one needs today is a laptop and an internet connection and a few hours, they can wreak havoc. In the past, you'd need TV or newspapers to spread bad (and good) stuff - they were expensive, time consuming to produce and had limited reach.
  • kloop 2y ago
    There are lots of organizations with $1M and a desire to influence the population This can only be done with a sentiment that was, at least partially, already there. And may very well happen naturally eventually
  • immibis 2y ago
    How can I wreck havoc with a few hours, a laptop, and an internet connection? It takes a bit more than that.
    • akudha 2y ago
      Some woman’s cat was hiding in her basement. She automatically assumed her Haitian neighbors stole her cat and made some comment about it, which landed on Facebook, which got morphed into “immigrants eating pets” story, JD Vance picked it up, Trump mentioned it in a national debate watched by 65 million people. All of this happened in a few days. This resulted in violence in Springfield. If you can place a rumor or lie in front of the right person/people to amplify, it will be amplified. It will spread like wildfire, and by the time it is fact checked, it will have done at least some damage.
      • immibis 2y ago
        These successful manipulation stories are extremely rare though. What usually happens is that you say your neighbour ate your cat, then everyone laughs at you. Did the person who posted do the manipulation, or did JD Vance and Donald Trump do it?
• heresie-dabord 2y ago
  > you trusted what you read before. For me, LLMs don't change anything. I already questioned the information before and continue to do so. [...] Why do you think that you could trust what you read before? A human communicator is, in a sense, testifying when communicating. Humans have skin in the social game. We try to educate people, we do want people to be well-informed and to think critically about what they read and hear. In the marketplace of information, we tend very strongly to trust non-delusional, non-hallucinating members of society. Human society is a social-confidence network. In social media, where there is a cloak of anonymity (or obscurity), people may behave very badly. But they are usually full of excuses when the cloak is torn away; they are usually remarkably contrite before a judge. A human communicator can face social, legal, and economic consequences for false testimony. Humans in a corporation, and the corporation itself, may be held accountable. They may allocate large sums of money to their defence, but reputation has value and their defence is not without social cost and monetary cost. It is literally less effort at every scale to consult a trusted and trustworthy source of information. It is literally more effort at every scale to feed oneself untrustworthy communication.
• sevensor 2y ago
  For me, the problem has gone from “figure out the author’s agenda” to “figure out whether this is a meaningful text at all,” because gibberish now looks a whole lot more like meaning than it used to.
  • pxoe 2y ago
    This has been a problem on the internet for the past decade if not more anyway, with all of the seo nonsense. If anything, maybe it's going to be ever so slightly more readable.
    • orthecreedence 2y ago
      I don't know what you're talking about. Most people don't think of SEO, Search Engine Optimization, Search Performance, Search Engine Relevance, Search Rankings, Result Page Optimization, or Result Performance when writing their Article, Articles, Internet Articles, News Articles, Current News, Press Release, or News Updates...
• solidninja 2y ago
  There's a quantity argument to be made here - before, it used to be hard to generate large amounts of plausible but incorrect text. Now it easy. Similar to surveillance before/after smartphones + the internet - you had to have a person following you vs just soaking up all the data on the backbone.
• escape_goat 2y ago
  There was a degree of proof of work involved. Text took human effort to create, and this roughly constrained the quantity and quality of misinforming text to the number of humans with motive to expend sufficient effort to misinform. Now superficially indistinguishable text can be created by an investment in flops, which are fungible. This means that the constraint on the amount of misinforming text instead scales with whatever money is resourced to the task of generating misinforming text. If misinforming text can generate value for someone that can be translated back into money, the generation of misinforming text can be scaled to saturation and full extraction of that value.
• low_tech_love 2y ago
  It’s nothing to do with trusting in terms of being true or false, but whatever I read before I felt like, well, it can be good or bad, I can judge it, but whatever it is, somebody wrote it. It’s their work. Now when I read something I just have absolutely no idea whether the person wrote it, how much percent did they write it, or how much they even had to think before publishing it. Anyone can simply publish a perfectly well-written piece of text about any topic whatsoever, and I just can’t wrap my head around why, but it feels like a complete waste of time to read anything. Like… it’s all just garbage, I don’t know.
• danielmarkbruce 2y ago
  The following appears to be true: If one spends a lot of years reading a lot of stuff, they come to this conclusion, that most of it cannot be trusted. But it takes lots of years and lots of material to see it. If they don't, they don't.
• mvdtnz 2y ago
  It's that you trusted that what you read came from a human being. Back in the day I used to spend hours reading Evolution vs Creationism debates online. I didn't "trust" the veracity of half of what I read, but that didn't mean I didn't want to read it. I liked reading it because it came from people. I would never want to read AI regurgitation of these arguments.