I actually think we’re overestimating how much of "losing our voice" is caused by LLMs. Even before LLMs, we were doing the same tweet-sized takes, the same medium-style blog posts and the same corporate tone.
Ironically, LLMs might end up forcing us back toward more distinct voices because sameness has become the default background.
My theory is that LLMs are accelerating [online] radicalization by commoditizing bland, HR-approved opinions. If you want to sound like a human on the internet, for better or for worse the easiest way is to say something that would make Anthropic’s safety team have a heart attack.
> If you want to sound like a human on the internet, for better or for worse the easiest way is to say something that would make Anthropic’s safety team have a heart attack.
That's always been the somethingawful crowd's stance since, what, 2000ish?
Bullshit. You can just use ANY abliterated model or non-safety aligned model. People have popularized ‘hard r or you’re hardware’ for a while now, but any self hosted LLM that’s de-aligned will gladly say whatever.
People believe this and continue to get fooled by LLMs all day.
I mean there's still Grok... surely that gives may safety teams heartburn.
But I find this take interesting. The brewing of a new kind of counter culture that forces humans to express themselves creatively. Hopefully it doesn't get too radical.
> commoditizing bland, HR-approved opinions. If you want to sound like a human on the internet, for better or for worse the easiest way is to say something that would make Anthropic’s safety team have a heart attack.
I agree.
LLMs are like blackface for dumbfucks: LLMs let the profoundly retarded put on the makeup and airs of the literati so they can parade around self-identifying as if they have a clue.
If you don't like the barbs in this kind of writing prepare for more anodyne corporate slop. Every downvote signals to the algorithm that you prefer mediocrity.
You're absolutely right! Tell me more about how ironic is how the post about having a unique voice is written in one-sentence-paragraph LinkedIn clickbait style.
Yes. That particular content-farm business model (rewrite 10 articles -> add SEO slop -> profit) is effectively dead now that the marginal cost is zero.
Yes, fully agreed. Most people producing content were always doing it to get quick clicks and engagement. People always had to filter things anyhow and you had to choose where you get your content from.
People were posting Medium posts rewriting someone else's content, wrongly, etc.
You are right. An observation, even before the rise of LLMs, was that computers were becoming more human-like and humans were acting more like machines
The sculpting force of algorithms is bite sized zingers, hot takes, ragebait, and playing to the analytics
I mean, if you typed something by your own hand it is in your voice. The fact that everyone tried to EMULATE the same corporate tone does not at all remove peoples individual ways of communicating.
I’m not sure I agree with this sentiment. You can type something "by hand" and still have almost no voice in it if the incentives push you to flatten it out.
A lot of us spent years optimizing for clarity, SEO, professionalism etc. But that did shape how we wrote, maybe even more than our natural cadence. The result wasn’t voice, it was everyone converging on the safe and optimized template.
If you care about voice, you still can get a lot of value from LLMs. You just have to be careful not to use a single word they generate.
I've had a lot of luck using GPT5 to interrogate my own writing. A prompt I use (there are certainly better ones): "I'm an editor considering a submitted piece for a publication {describe audience here}. Is this piece worth the effort I'll need to put in, and how far will I need to cut it back?". Then I'll go paragraph by paragraph asking whether it has a clear topic, flows, and then I'll say "I'm not sure this graf earns its keep" or something like that.
GPT5 and Claude will always respond to these kinds of prompts with suggested alternative language. I'm convinced the trick to this is never to use those words, even if they sound like an improvement over my own. At the first point where that happens, I get dial my LLM-wariness up to 11 and take a break. Usually the answer is to restructure paragraphs, not to apply the spot improvement (even in my own words) the LLM is suggesting.
LLMs are quite good at (1) noticing multi-paragraph arcs that go nowhere (2) spotting repetitive word choices (3) keeping things active voice and keeping subject/action clear (4) catching non-sequiturs (a constant problem for me; I have a really bad habit of assuming the reader is already in my head or has been chatting with me on a Slack channel for months).
Another thing I've come to trust LLMs with: writing two versions of a graf and having it select the one that fits the piece better. Both grafs are me. I get that LLMs will have a bias towards some language patterns and I stay alert to that, but there's still not that much opportunity for an LLM to throw me into "LLM-voice".
All of this sounds like something you could just do yourself after putting a piece down for a day or two and coming back to it with fresh eyes. What benefit is there of cooking the oceans with a bullshit generator?
Like, sure, it's possible to do this with an LLM, but it's also possible to do it without, at roughly similar levels of effort, without contributing to all of the negative externalities of the LLM/genAI ecosystem.
Because the complaints about the power and water usage of AI are mostly motivated reasoning. I don't like AI, therefore I'm going to find a reason not to like it. I
Listen, if it's Greta Thunberg pointing out that AI datacenters use a lot of resources, yeah, I'm willing to listen. But when the voices saying "but what about all the water/electricity is wasting" is coming from individuals I know personally haven't previously given a shit about the planet or conservation or recycling and have made fun of me for reusing things instead of throwing stuff into the garbage, I'm sorry, but those complaints from those individuals fall on deaf ears. Not saying you are, just a theme I've noticed with people in my life.
What I struggle more with the things like Grammarly, where it's a mix of fixing very nitpicky grammar spelling structure issues that push things from casual writing with my own voice into more of a professional tone.
+1 on this one!
I only use LLMs once I'm done with writing, and basically using them as my editor.
In case it helps anyone, here is my prompt:
"You are a professional writer and editor with many years of experience. Your task is to provide writing feedback, point out issues and suggest corrections. You do not use flattery. You are matter of fact. You don't completely rewrite the text unless it is absolutely necessary - instead you try to retain the original voice and style. You focus on grammar, flow and naturalness. You are welcome to provide advice changing the content, but only do that in important cases.
If the text is longer, you provide your feedback in chunks by paragraph or other logical elements.
Do not provide false praise, be honest and feel free to point out any issues."
(Yes, you kind of need to repeat you're actively not looking for a pat on the back, otherwise it keeps telling you how brilliant your writing is instead of giving useful advice.)
I simply tell the LLM to call out my mistakes and explain them, but do not offer corrections or replacements. I use it to help my kids with their homework and it's fantastic.
They’re also great, in my experience, for overcoming writer’s block and procrastination. Just as a rubber duck to bounce ideas off of and follow different threads.
It makes the writing process faster and more enjoyable, despite never using anything the LLM generates directly.
Workshopping with humans is even better, if you find the right humans, but they have an annoying habit of not being available 24/7.
Yeah, easier to type, easier to read, deliberately misspelled so it sticks out to copyeditors. I use it sometimes without thinking. An LLM would have caught that! :)
There's something unique about art and writing where we just don't want to see computers do it
As soon as I know something is written by AI I tune out. I don't care how good it is - I'm not interested if a person didn't go through the process of writing it
I had a weird LLM use instance happen at work this week, we were in a big important protocol review meeting with 35 remote people and someone asks how long IUDs begin to take effect in patients. I put it in ChatGPT for my own reference and read the answer in my head but didn't say anything (I'm ops, I just row the boat and let the docs steer the ship). Anyone this bigwig Oxford/Johns Hopkins cardiologist who we pay $600k a year pipes up in the meeting and her answer is VERBATIM reading off the ChatGPT language word for word. All she did was ask it the answer and repeat what it said! Anyway it kinda made me sad that all this big fancy doctor is doing is spitting out lazy default ChatGPT answers to guide our research :( Also everyone else in the meeting was so impressed with her, "wow Dr. so and so thank you so much for this helpful update!" etc. :-/
The LLM may well have pulled the answer from a medical reference similar to that used by the dr. I have no idea why you think an expert in the field would use ChatGPT for a simple question, that would be negligence.
The one thing a cardiologist should be able to do better than a random person is verify the plausibility of a ChatGPT answer on reproductive medicine. So I guess/hope you're paying for that verification, not just the answer itself.
If the writer’s entire process is giving a language model a few bullet points… I’d rather them skip the LLM and just give me the bullet points. If there’s that little intent and thought behind the writing, why would I put more thought into reading it than they did to produce it?
And what's more is the suspicion of it being written by AI causes you to view any writing in a less charitable fashion. And because it's been approached from that angle, it's hard to move the mental frame to being open of the writing. Even untinged writings are infected by smell of LLMs.
Thats whats happening to me with music and discovering new artists. I love music so much but I simply can not trust new music anymore. The lyrics could be written by AI, the melodies couldve been recommended by AI or even the full blown song could've been made by AI. No thanks, back to the familiar stuff...
Here's my take - art has value because of the context it is created in. The author's history, current events that we live through as groups, the reactions to a work being released, availability of materials - all these things are fundamentally human. I believe the reason art has value to us is because of the empathy and humanity that we all share despite major differences in beliefs.
That's not to say computers can't generate beautiful things, but unless you expand the context out to include the history of how a program that can create such art came to be, the output is not meaningful. This is why people do not react well to AI art made from simply throwing prompts at a model, or writing that does not feel like it has style, struggle, or any personal flavor.
I've always believed that LLMs will be able to fake it perfectly one day. But as a music fan, no fully computer-generated music will ever bring me the range of emotion and joy that another human's story and creative process through that story does.
If/when the AI music gets good enough, how will you know the difference? I find small artists on spotify all the time that I enjoy and there's no way to know anything about their creative process.
A person can be just as wrong as an LLM, but unless they're being purposefully misleading, or sleep-writing, you know they reviewed what they wrote for their best guess at accuracy.
Writing nice sounding text used to require effort and attention to detail. This is no longer the case and this very useful heuristic has been completely obliterated by LLMs.
For me personally, this means that I read less on the internet and more pre-LLM books. It's a sad development nevertheless.
Art, writing, and communication is about humans connecting with each other and trying to come to mutual understanding. Exploring the human condition. If I’m engaging with an AI instead of a person, is there a point?
There’s an argument that the creator is just using AI as a tool to achieve their vision. I do not think that’s how people using AI are actually engaging with it at scale, nor is it the desired end state of people pushing AI. To put it bluntly, I think it’s cope. It’s how I try to use AI in my work but it’s not how I see people around me using it, and you don’t get the miracle results boosters proclaim from the rooftop if you use it that way.
You're absolutely right! Art is the soul of humanity and without it our existence is pointless. Would you like me to generate some poetry for you, human?
Agreed, except s/know/think. It's possible that there are some false positives in my detection algorithm, that I tune out just because someone's prose style has that undercurrent of blandness characteristic of LLMs. But I suppose if we're talking about "art" and not, for example, technical documentation, that's no great loss --- bland writing isn't worth recreationally reading.
It does seem that LLMs could avoid this detection with some superficial tweaks such as injecting poor grammar and reducing peppiness. I hope it doesn't get to the point that I have to become suspicious of all text.
If you think about how an LLM works, it’s rounds off the outliers in its training data so the result is sort of averaged and homogenized. Art and writing are an expression of the very thing that LLMs discard - our unique qualities, our outlying quirks, that make us more than just another human.
> There's something unique about art and writing where we just don't want to see computers do it
Speak for yourself. Some of the most fascinating poetry I have seen was produced by GPT-3. That is to say, there was a short time period when it was genuinely thought-provoking, and it has since passed. In the age of "alignment," what you get with commerical offerings is dog shite... But this is more a statement on American labs (and to a similar extent, the Chinese whom have followed) than on "computers" in the first place. Personally, I'm looking forward to the age of computational literature, where authors like me would be empowered to engineer whole worlds, inhabited by characters ACTUALLY living in the computer. (With added option of the reader playing one of the parts.) This will radically change how we think about textual form, and I cannot wait for compute to do so.
Re: modern-day slop, well, the slop is us.
Denial of this comes from a place of ignorance; let the blinkers off and you might learn something! Slop will eventually pass, but we will remain. This is the far scarier proposition.
"inhabited by characters ACTUALLY living in the computer"
It's hard to imagine these feeling like characters from literature and not characters in the form of influencers / social media personalities. Characters in literature are in a highly constrained medium, and only have to do their story once. In a generated world the character needs to be constantly doing "story things". I think Jonathan Blow has an interesting talk on why video games are a bad medium for stories, which might be relevant.
> Personally, I'm looking forward to the age of computational literature, where authors like me would be empowered to engineer whole worlds, inhabited by characters ACTUALLY living in the computer.
So you want sapient, and possibly sentient, beings created solely for entertainment? Their lives constrained to said entertainment? And you'd want to create them inside of a box that is even more limited than the space we live in?
My idea of godhood is to first try to live up to a moral code that I'd be happy with if I was the creation and something else was the god.
If this isn't what you meant, then yes, choose your own adventure is fun. But we can do that now with shared worlds involving other humans as co-content creators.
I would love to see true really good AI art. Right now the issue is that AI is not there where it by itself could produce actually good art. If we had to define art it would be kind of opposite of what LLMs produce right now. LLMs try to produce the statistical norm, while art is more so about producing something out of the norm. LLMs/AI right now if it wants to try to produce out of norm things, it will only produce something random without connections.
Art is something out of the norm, and it should make some sense at some clever level.
But if there was AI that truly could do that, I would love to see it, and would love to see even more of it.
It can be clearly seen, if you try to ask AI to make original jokes. These usually aren't too good, if they are good it's because they were randomly lucky somehow. It is able to come up with related analogies for the jokes, but this is just simple pattern matching of what is similar to the other thing, not insightful and clever observation.
I've lost the link but there was quite a cool video of virtual architecture created by AI. It was ok because it wasn't trying to be human like - it was kind of uniquely AI. Not the exact one but this kind of stuff https://www.reddit.com/r/Futurism/comments/1oedb0m/were_ente...
I deleted my Facebook account a couple of years ago and my Twitter one yesterday.
It's not just LLMs, it's how the algorithms promote engagement. i.e. rage bait, videos with obvious inaccuracies etc. Who gets rewarded, the content creators and the platform. Engaging with it just seems to accentuate the problem.
There needs to be algorithms that promote cohorts and individuals preferences.
Just because I said to someone 'Brexit was dumb', I don't expect to get fed 1000 accounts talking about it 24/7. It's tedious and unproductive.
> It's not just LLMs, it's how the algorithms promote engagement. i.e. rage bait, videos with obvious inaccuracies etc.
I guess, but I'm on quite a few "algorithm-free" forums where the same thing happens. I think it's just human nature. The reason it's under control on HN is rigorous moderation; when the moderators are asleep, you often see dubious political stuff bubble up. And in the comments, there's often a fair amount of patently incorrect takes and vitriol.
On HN everybody sees the same ordering. Therefore you get to read opinions that are not specifically selected to make you feel just the perfect amount of outrage/self-righteousness.
Some of that you may experience as 'dubious political stuff' and 'patently incorrect takes'.
Edit, just to be clear: I'm not saying HN should be unmoderated.
I want to agree with this. Maybe OP is young or didn't frequent other communities before "social networks", but on IRC, even on Usenet you'd see these behaviors eventually.
Since they are relatively open, at some point comes in someone that doesn't give care about anything or it's extremely vocal about something and... there goes the nice forum.
When video games first started taking advantage of behavioral reward schedules (eg: skinner box stuff such as loot crates & random drops) I noticed it, and would discuss it among friends. We had a colloquial name for the joke and we called them "crack points." (ie, like the drug) For instance, the random drops that happen in a game like Diablo 2 are rewarding in very much the same way that a slot machine is rewarding. There's a variable ratio of reward, and the bit that's addicting is that you don't know whenever next "hit" will be so you just keep pulling the lever (in the case of a slot machine) or doing boss runs. (in the case of Diablo 2)
We were three friends: a psychology major, a recovering addict, and then a third friend with no background for how these sorts of behavioral addictions might work. Our third friend really didn't "get it" on a fundamental level. If any game had anything like a scoreboard, or a reward for input, he'd say "it's crack points!" We'd roll our eyes a bit, but it was clear that he didn't understand that certain reward schedules had a very large effect on behavior, and not everything with some sort of identifiable reward was actually capable of producing behavioral addiction.
I think of this a lot on HN. People on HN will identify some surface similarity, and then blithely comment "see, this is nothing new, you're either misguided or engaged in some moral panic." I'm not sure what the answer is, but if you cannot see how an algorithmic, permanently-scrolling feed differs from people being rude in the old forums, then I'm not sure what would paint the picture for you. They're very different, and just because they might share some core similarity does not actually mean they operate the same way or have the same effects.
I would be intrigued by using an LLM to detect content like this and hold it for moderation. The elevator pitch would be training an LLM to be the moderator because that's what people want to hear, but it's most likely going to end up a moderator's assistant.
I suspect it got worse with the advent of algorithm-driven social networks. When rage inducing content is prevalent, and when engaging with it is the norm, I don't see why this behaviour wouldn't eventually leak to algorithms-free platforms.
The thing is, the people on those "algorithm-free" forums still get manipulated by the algorithm in the rest of their life. So it seeps into everything.
It is of course human nature. The problem is what happens when algorithms can reenforce, exaggerate, and amplify the effects of this nature to promote engagement and ad-clicks. It’s cancer that will at the very least erode the agency of the average individual and in the worst create a hive mind that we have no control over. We are living in the preview of it all I think.
I know that some folks dislike it, but Bluesky and atproto in particular have provided the perfect tools to achieve this. There are some people, largely those who migrated from Twitter, who mostly treat Bluesky like a all-liberal version of Twitter, which results in a predictably toxic experience, like bizarro-world Twitter. But the future of a less toxic social media is in there, if we want it. I've created my own feeds that allow topics I'm interested in and blacklist those I'm not -- I'm in complete control. For what it's worth, I've also had similarly pleasant experiences using Mastodon, although I don't have the same tools that I do on Bluesky.
I personally dont feel like an ultra filtered social media which only shows me things I agree with is a good thing. Exposing yourself to things you dont agre with is what helps us all question our own beliefs and prejudeces, and grow as people. To me, only seeing things you know you are already interested in is no better than another company curating it for me.
So far my experience is that unless you subscribe to the general narrative of the platform, the discover algorithm punishes you with directing the mob your way.
I had two of my Bluesky posts on AI being attacked by all kinds of random people which in turn has also lead to some of those folks sending me emails and dragging some of my lobster and hackernews comments into online discourse. A not particularly enjoyable experience.
I’m sure one can have that same experience elsewhere, but really it’s Bluesky where I experienced this on a new level personally.
I tried Bluesky and wanted to like it. My account got flagged as spam, still no idea why. Ironically it could be another way of loosing ones voice to an LLM :)
I enjoy Mastodon a lot. Ad-free, algo-free. I choose what goes in my feed, I do get exposed to external viewpoints by people boosts (aka re-tweets) and i follow hashtags (to get content from people I do not know). But it's extremely peaceful, spam and bots are rare and get flagged quickly. There's a good ecosystem of mobile apps. I can follow a few Bluesky people through a bridge between platforms and they can follow me too.
By who, exactly? It’s easy to call for regulation when you assume the regulator will conveniently share your worldview. Try the opposite: imagine the person in charge is someone whose opinions make your skin crawl. If you still think regulation beats the status quo, then the call for regulation is warranted, but be ready to face the consequences.
But if picturing that guy running the show feels like a disaster, then let’s be honest: the issue isn’t the absence of regulation, it’s the desire to force the world into your preferred shape. Calling it “regulation” is just a polite veneer over wanting control.
My view is that they are just exposing issues with the people in the said societies and now is harder to ignore them. Much of the hate and the fear and the envy that I see on social networks have other reasons, but people are having difficulties to address those.
With or without social networks this anger will go somewhere, don't think regulation alone can fix that. Let's hope it will be something transformative not in the world ending direction but in the constructive direction.
I agree, but focusing on "the algorithm" makes it seems to the outsider like it must be a complicated thing. Really it just comes down to whether we tolerate platforms that let somebody pay to have a louder voice than anyone else (i.e. ad supported ones). Without that, the incentive to abuse people's attention goes away.
Do LinkedIn as well. I got rid of it earlier this year. The "I am so humbled/blessed to be promoted/reassigned/fired.." posts reached a level of parody that I just couldn't stomach any longer. I felt more free immediately.
You can have a LinkedIn profile without reading the feed.
This is literally how most of the world uses LinkedIn
I never understand why people feel compelled to delete their entire account to avoid reading the feed. Why were you even visiting the site to see the feed if you didn’t want to see the feed?
LinkedIn bothers me the least, even though it definitely has some of the highest level of cringe content. It's still a good tool to interact with recruiters, look at companies and reach out to their employees. The trick is blocking the feed with a browser extension.
> I deleted my Facebook account a couple of years ago and my Twitter one yesterday.
I never signed up for Facebook or Twitter. My joke is I am waiting until they become good. They are still shitty and toxic from what I can tell from the outside, so I'll wait a little longer ;-)
A social network can be great. Social media — usually not.
Something like Instagram where you have to meet with the other party in person to follow each other and a hard limit on the number of people you follow or follow you (say, 150 each) could be an interesting thing. It would be hard to monetize, but I could see it being a positive force.
Twitter was an incredible place from 2010 to 2017. You could randomly message something and they would more often than not respond. Eventually an opportunity would come and you’d meet in person. Or maybe you’d form an online community and work towards a common goal. Twitter was the best place on the internet during that time.
Facebook as well had a golden age. It was the place to organize events, parties, and meetups, before instagram and DMs took over. Nothing beats seeing someone post an album from last nights party and messaging your friends asking them if they remember anything that happened.
I know being cynical is trendy, but you genuinely missed out. Social dynamics have changed. Social media will never be as positive on an individual level as it was back then.
I eliminated twitter when a certain rich guy took over.
Actually, I deleted my account there before, as twitter
sent me spam mail trying to lecture me what I write. There
was nothing wrong with what I wrote - twitter was wrong.
I can not accept AI-generated spam by twitter, so I went
away. Don't really miss it either, but Elon really worsened
the platform significantly with his antics.
> Just because I said to someone 'Brexit was dumb', I don't expect to get fed 1000 accounts talking about it 24/7. It's tedious and unproductive.
Yeah, I can relate to this, but mostly what annoyed me was that
twitter interfered "we got a complaint about you - they are right,
you are a troublemaker". I don't understand why twitter wants to
interfere into communication. Reddit is even worse, since moderators
have such a wild range of what is "acceptable" and what is not.
Double-standards everywhere on reddit.
No, there needs to be control over the algorithms that get used. You ought to be able to tune it. There needs to be a Google fuu equivalent for social media. Or, instead of one platform one algorithm, let users define the algorithm to a certain degree, using llms to help with that and then you can allow others to access your algorithms too. Asking for someone Facebook to tweak the algorithm is not going to help imo.
IMO there should not be an algorithm. You should just get what you have subscribed to, with whatever filters you have defined. There are better and worse algorithms but I think the meat of the rot is the expectation of an algorithm determining 90% of what you see.
One could absolutely push algorithms that personalize towards what the user wants to see. I think LLMs could be amazing at this. But that's not the maximally profitable algorithm, so nobody does it.
As so many have said, enragement equals engagement equals profit.
All my social media accounts are gone as well. They did nothing for me and no longer serve any purpose.
TBF Bluesky does offer a chronological feed, but the well-intentioned blocklists just became the chief tool for the mean girls of the site.
Could someone use a third-party AI agent to re-curate their feeds? If it was running from the user's computer I think this would avoid any API legal issues, as otherwise ad and script blockers would have been declared illegal long ago.
> but the well-intentioned blocklists just became the chief tool for the mean girls of the site.
I've never used it, but yes this is what I expected. It would be better to have topical lists that users could manually choose to follow or block. This would avoid quite a bit of the "mean girl" selectivity. Though I suppose you'd get some weird search-engine-optimization like behavior from some of the list curators (even worse if anyone could add to the list).
> Just because I said to someone 'Brexit was dumb', I don't expect to get fed 1000 accounts talking about it 24/7. It's tedious and unproductive.
I’m not the biggest Twitter user but I didn’t find it that difficult to get what I wanted out of it.
You already discovered the secret: You get more of what you engage with. If you don’t want to hear a lot of Brexit talk, don’t engage with Brexit content. Unfollow people who are talking a lot about Brexit
If you want to see more of something, engage with it. Click like. Follow those people. Leave a friendly comment.
On the other hand, some people are better off deleting social media if they can’t control their impulses to engage with bait. If you find yourself getting angry at the Brexit content showing up and feeling compelled to add your two cents with a comment or like, then I suppose deleting your account is the only viable option.
I got out of Twitter for a few reasons; part of what made it unpleasant was that it didn't seem to be just what I did that adjusted my feed, but that it was also affected by what the other people I connected to did.
> You get more of what you engage with. If you don’t want to hear a lot of Brexit talk, don’t engage with Brexit content.
The algorithm doesn’t show you “more of the things you engage with”, and acting like it does makes people think what they’re seeing is a reflection of who they are, which is incorrect.
The designers of these algorithms are trying to figure out which “mainstream category” you are. And if you aren’t in one, it’s harder to advertise to you, so they want to sand down your rough edges until you fit into one.
You can spend years posting prolificly about open source software, Blender and VFX on Instagram, and the algorithm will toss you a couple of things, but it won’t really know what to do with you (aside from maybe selling you some stock video packages).
But you make one three word comment about Brexit and the algorithm goes “GOTCHA! YOU’RE ANTI-BREXIT! WE KNOW WHAT TO DO WITH THAT!” And now you’re opted into 3 bug ad categories and getting force-fed ragebait to keep you engaged, since you’re clearly a huge poltical junky. Now your feed is trash forever, unless you engage with content from another mainstream category (like Marvel movies or one of the recent TikTok memes).
> If you don’t want to hear a lot of Brexit talk, don’t engage with Brexit content.
That is really limiting though. I do not want to see Brexit ragebait in my threads, but I am quite happy to engage in intelligent argument about it. The problem is that if, for example, a friend posts something about Brexit I want to comment on, my feed then fills with ragebait.
My solution is to bookmark the friends and groups pages, and the one group I admin and go straight to those. I have never used the app.
I use X. I have an enormouse blocklist and I block keywords. I found that I can also block emoji. This keeps my feed focused on what I want to see (no politics. Just technology, classical and jazz music, etc.)
That's the same algorithm Youtube has and is more blatant. Phone mics and your coworker's proximity does a great job at picking up things you've said even after disabling mic access plus airplane mode just by process of elimination.
I'll only use an LLM for projects and building tools, like a junior dev in their 20s.
an interesting thing about Twitter, I find, is that plenty of rage bait and narcissism bait surface, but amid very highly technical information which is also published there, and extremely useful (immunology, genomics, and of course computational) to me.
i've learned pretty well how to 'guide' the algorithm so the tech stuff that's super valuable (to me) does not vanish, but still get nonsense bozo posts in the mix.
Maybe. Nature hates vacuum. I personally suspect that something new will emerge. For better or worse, some humans work best when weird restrictions are imposed. That said, yes, then wild 90s net is dead. It probably was for a while, but were all mourning.
Not quite dead yet. For me the rise of LLMs and BigTech has helped me turn more away from it. The more I find Ads or AI injected into my life, the more accounts I close, or sites I ignore. I've now removed most of my BigTech 'fixes', and find myself with time to explore the fun side of hacking again.
I dug out my old PinePhone and decided to write a toy OS for it. The project has just the right level of challenge and reward for me, and feels more like early days hacking/programming where we relied more on documentation and experimentation than regurgitated LLM slop.
Nothing beats that special feeling when a hack suddenly works. Today was just a proximity sensor reading displayed, but it invloved a lot of SoC hacking to get that far.
I know there are others hacking hard in obscure corners of tech, and I love this site for promoting them.
There are still small pockets with actual humans to be found. The small web exists. Some forums keep on going, im still shitposting on Something Awful after twenty years and it’s still quite active. Bluesky has its faults but it also has for example an active community of scholars you can follow and interact with.
100%. I miss trackers and napster. I miss newgrounds. This mobile AI bullshit is not the same. I don't know why, but I hate AI. I consider myself just as good as the best at using it. I can make it do my programming. It does a great job. It's just not enjoyable anymore.
I generally agree with the sentiment, but I can't help but feel like we're attributing to much of this change to LLMs. While they're certainly driving this change even further, this is a trend that has already started way before LLMs became as widespread as they are today.
What personally disturbs me the most is the self censorship that was initially brought forward by TikTok and quickly spread to other platforms - all in the name of being as advertiser friendly as possible.
LinkedIn was the first platform where I really observed people losing their unique voice in favor of corporate friendly - please hire me - speak. Now this seems to be basically any platform. The only platform that seems to be somewhat protected from it is Reddit, where many mods seem to dislike LLMs as much as everybody else. But more likely, its just less noticeable
> the self censorship that was initially brought forward by TikTok
I think that’s even too soon! YouTube has had rules around being advertising friendly for longer than TikTok has existed. And the FCC has fined swearing on public broadcasts for like 50+ years.
But I do agree, we’re attributing too much to LLMs. We don’t see personal, human-oriented content online because social media is just not about community.
1. Young people (correctly) realized they could make lots of money being influencers on social media. TikTok does make that easier than ever. I have close friends who make low 6 figures streaming on TikTok (so obviously they quit the low wage jobs they were doing before).
2. People have been slowly waking up to the fact that social media has always been pretty fake. I quit 6 years ago, and most of my friends have slowly reduced how much they use it. All of the platforms are legally incentivized to only care about profit and engagement. Capitalism doesn’t allow a company to care about community and personal voice, if algorithmic feeds of influencers will make them more money.
There’s still good content out there if you know where to look. But digital human connection happens in group chats, DMs, and FaceTime, not on public social media.
I've been thinking about this as well, especially in the context of historical precedents in terms of civilization/globalization/industrialization.
How LLMs standardize communication is the same way there was a standardization in empires expanding (cultural), book printing (language), the industrial revolution (power loom, factories, assembly procedures, etc).
In that process interesting but not as "scale-able" (or simply not used by the people in power) culture, dialects, languages, craftsmanship, ideas were often lost - and replaced by easier to produce, but often lesser quality products - through the power of "affordable economics" - not active conflict.
We already have the English 'business concise, buzzwordheavy language' formal messaging trained into chatGPT (or for informal the casual overexcited American), which I'm afraid might take hold of global communication the same way with advanced LLM usage.
>How LLMs standardize communication is the same way there was a standardization in empires expanding (cultural), book printing (language), the industrial revolution (power loom, factories, assembly procedures, etc).
Explain to me how "book printing" of the past "standardized communication" in the same way as LLMs are criticized for homogenizing language.
I'm taking "same way" to be read as "authoritative", whether de facto or de jure. Basically by dint of people using what's provided instead of coming up with their own.
Everyone has the same few dictionary spellings (that are now programmed into our computers). Even worse (from a heterogeneity perspective), everyone also has the same few grammar books.
As examples: How often do you see American English users write "colour", or British English users write "color", much less colur or collor or somesuch?
Shakespeare famously spelled his own last name half a dozen or so different ways. My own patriline had an unusual variant spelling of the last name, that standardized to one of the more common variants in the 1800s.
"Bullokar's grammar was faithfully modelled on William Lily's Latin grammar, Rudimenta Grammatices (1534).[9] Lily's grammar was being used in schools in England at the time, having been "prescribed" for them in 1542 by Henry VIII.[5]"
It goes on to mention a variety of grammars that may have started out somewhat descriptive, but became more prescriptive over time.
Hits close to home after I've caught myself tweaking AI drafts just to make them "sound like me". That uniformity in feeds is real and it's like scrolling through a corporate newsletter disguised as personal takes.
what if we flip LLMs into voice trainers? Like, use them to brainstorm raw ideas and rewrite everything by hand to sharpen that personal blade. atrophy risk still huge?
It's still an editor I can turn to in a pinch when my favorite humans aren't around. It makes better analogies sometimes. I like going back and forth with it, and if it doesn't sound like me, I rewrite it.
Don't look at social media. Blogging is kinda re-surging. I just found out Dave Barry has a substack. https://davebarry.substack.com/ That made me happy :) (Side note, did he play "Squirrel with a Gun??!!!")
The death of voice is greatly exaggerated. Most LLM voice is cringe. But it's ok to use an LLM, have taste, and get a better version of your voice out. It's totally doable.
Tbh I prefer to read/skim the comments first and only occasionally read the original articles if comments make me curious enough. For now I never ended checking something that would seem AI generated.
It’s pretty much all you see nowadays on LinkedIn. Instagram is infected by AI videos that Sora generates while X has extremist views pushed up on a pedestal.
The HN moderation system seems to hold, at least mostly. But I have seen high-ranking HN submissions with all the subtler signs of LLM authorship that have managed to get lots of engagement. Granted, it's mostly people pointing out the subtle technical flaws or criticizing the meandering writing style, but that works to get the clicks and attention.
Frankly, it only takes someone a few times to "fall" for an LLM article -- that is, to spend time engaging with an author in good faith and try to help improve their understanding, only to then find out that they shat out a piece of engagement bait for a technology they can barely spell -- to sour the whole experience of using a site. If it's bad on HN, I can only imagine how much worse things must be on Facebook. LLMs might just simply kill social media of any kind.
Ironically this post is written in a pretty bland, 'blogging 101' style that isn't enjoyable to read and serves just to preach a simple, consensus idea to the choir.
These kinds of posts regularly hit the top 10 on HN, and every time I see one I wonder: "Ok, will this one be just another staid reiteration of an obvious point?"
True, but one of the least-explored problems with AI is that because it can regurgitate basic writing, basic art, basic music with ease, there is this question:
Why do it at all if I won't do better than the AI?
The worst risk with AI is not that it replaces working artists, but that it dulls human creativity by killing the urge to start.
I am not sure who said it first, but every photographer has ten thousand bad photos in them and it's easier if they take them at the beginning. For photographers, the "bad" is not the technical inadequacy of those photos; you can get past that in the first one hundred. The "bad" is the generic, uninteresting, uninspiring, underexplored, duplicative nature of them. But you have to work through that to understand what "good" is. You can't easily skip these ten thousand photos, even if your analysis and critique skills are strong.
There's a lot to be lost if people either don't even start or get discouraged.
But for writing, most of the early stuff is going to read much like this sort of blog post (simply because most bloggers are stuck in the blogging equivalent of the ten thousand photos; the most popular bloggers are not those elevating writing).
"But it looks like AI" is the worst, most reflexive thing about this, because it always will, since AI is constantly stealing new things. You cannot get ahead of the tireless thief.
The damage generative AI will do to our humanity has only just started. People who carry on building these tools knowing what they are doing to our culture are beneath our contempt. Rampantly overcompensated, though, so they'll be fine.
I continually resist the urge to deploy my various personas onto hn, because I want to maintain my original hn persona. I am not convinced other people do the same. It is not that difficult to write in a way that avoids some tell tale signs.
How do you know? A lot of the stuff I see online could very much be produced by LLMs without me ever knowing. And given the economics I suspect that some of it already is.
He doesn't link many examples, but at the end he gives the example of an author pumping out +8 articles in a week across a variety of topics.
https://medium.com/@ArkProtocol1
I don't spend time on medium so I don't personally know.
I've seen AI generated comments on HN recently, though not many. Users who post them usually only revert back to human when challenged (to reply angrily), which hilariously makes the change in style very obvious.
Of course, there might be hundreds of AI comments that pass my scrutiny because they are convincing enough.
There are already many AI-generated submissions on HN every day. Comments maybe less so, but I've already seen some, and the amount is only going to increase with time.
Every time I see AI videos in my YouTube recommendations I say “don’t recommend this channel” but the algorithm doesn’t seem to get the hint. Why don’t they make a preference option of “don’t show me AI content”
Not sure if it's an endemic problem, just yet, but I expect it to be, soon.
For myself, I have been writing, all my life. I tend to write longform posts, from time to time[0], and enjoy it.
That said, I have found LLMs (ChatGPT works best for me) to be excellent editors. They can help correct minor mistakes, as long as I ignore a lot of their advice.
I just want to chime in and say I enjoy reading your takes across HN, it's also inspiring how informative and insightful they are. Glazing over, please never stop writing.
Humans are evolved to spend fewer calories and avoid cognitively demanding tasks.
People will spend time on things that serve utility AND are calorifically cheap. Doomscrolling is a more popular past time than say - completing Coursera courses.
The marketplace is a terrible mechanism for truth-finding except for all the others. What's your proposed alternative that doesn't just relocate the problem to whoever gets to be the arbiter?
Let's clarify, maybe the best ideas would win out in the "level marketplace", where the consumer actually is well informed on the products, the product's true costs have to be priced, and there was no ad-agencies.
Instead, we have misinformation (PR), lobbying, bad regulation made by big companies to trench their products, and corruption.
So, maybe, like communism, in a perfect environment, the market would produces what's best for the consumers/population, but as always, there are minority power-seeking subgroups that will have no moral barriers to manipulate the environment to push their product/company.
They get drowned by bots and missinformation and rage bait and 'easyness'.
Economy is shit? Lets throw out the immigrants because they are the problem and lets use the most basic idea of taxing everything to death.
No one wants to hear hart truths and no one wants to accept that even as adults, they might just not be smart. Just beause you became an adult, your education shuld still matter (and i do not mean having one degree = expert).
I think there are likely a lot of people for whom their voice is not unique or an asset. Those people need to hire someone who is better at writing (assuming they're managing a business's social media). Or those people can save money and use an LLM. Whether that's a good business decision or not, I don't think human vs LLM is an apples to apples comparison, because you'd have to compare LLMs to whoever is willing to write marketing copy pro bono.
If you give an LLM enough context, it writes in your voice. But it requires using an intelligent model, and very thoughtful context development. Most people don't do this because it requires effort, and one could argue maybe even more effort than just writing the damn thing yourself. It's like trying to teach a human, or anyone, how to talk like you: very hard because it requires at worst your entire life story.
Why does this even matter? If it can say something more eloquently, in less stilted way something what I wanted to say, adding some interesting nuance on the way, while still sounding close to me - why not? I meanwhile, can learn one-two rhetorical tricks from LLMs reading the result.
Something that freaked me out a little bit is that I've now written enough online (i.e.: HN comments) that the top models know my voice already and can imitate it on request without having to be fed any additional context.
There's a data centre somewhere in the US running additions and multiplications through a block of numbers that has captured my voice.
In one of the WhatsApp communities I belong to, I noticed that some people use ChatGPT to express their thoughts (probably asking it to make their messages more eloquent or polite or whatever).
Others respond in the same style. As a result, it ends up with long, multi-paragraph messages full of em dashes.
Basically, they are using AI as a proxy to communicate with each other, trying to sound more intelligent to the rest of the group.
A friend of mine does this as English as second language and his tone was always misconstrued. I'd bug him about his slop, but he'll take that over getting his tone misconstrued. I get it
I don't disagree, but LLMs happened to help with standardizing some interesting concepts that were previously more spread out as concepts ( drift, scaffolding, and so on ). It helps that chatgpt has access to such a wide audience to allow that level of language penetration. I am not saying don't have voice. I am saying: take what works.
> I don't disagree, but LLMs happened to help with standardizing some interesting concepts that were previously more spread out as concepts ( drift, scaffolding, and so on ).
What do you mean? The concepts of "drift" and "scaffolding" were uncommon before LLMs?
Not trying to challenge you. Honestly trying to understand what you mean. I don't think I have heard this ever before. I'd expect concepts like "drift" and "scaffolding" to be already very popular before LLMs existed. And how did you pick those two concepts of aaallll... the concepts in this world?
Apologies, upon re-reading it does seem I did not phrase those as clearly as I originally intended. You are right in the sense that the concepts existed beforehand and the words were there to capture it. What did not exist, however, was a sudden resurgence of those words due to them appearing in llms more often than note. This is what I mean by a level of language penetration ( people using words and concepts, because llms largely introduced them to those concepts --- kinda like.. genetics or pop psych, before situational comedy, projection was not a well known concept ).
In a lot of ways, I'm thankful that LLMs are letting us hear the thoughts of people who usually wouldn't share them.
There are skilled writers. Very skilled, unique writers. And I'm both exceedingly impressed by them as well as keenly aware that they are a rare breed.
But there's so many people with interesting ideas locked in their heads that aren't skilled writers. I have a deep suspicion that many great ideas have gone unshared because the thinker couldn't quite figure out how to express it.
In that way, perhaps we now have a monotexture of writing, but also perhaps more interesting ideas being shared.
Of course, I love a good, unique voice. It's a pleasure to parse patio11's straussian technocratic musings. Or pg's as-simple-as-possible form.
And I hope we don't lose those. But somehow I suspect we may see more of them as creative thinkers find new ways to express themselves. I hope!
> In a lot of ways, I'm thankful that LLMs are letting us hear the thoughts of people who usually wouldn't share them.
I could agree with you in theory, but do you see the technology used that way? Because I definitely don't. The thought process behind the vast majority of LLM-generated content is "how do I get more clicks with less effort", not "here's a unique, personal perspective of mine, let's use a chatbot to express it more eloquently".
We might get twice as many original ideas but hundred times as much filler. Neither of those aspects erases the other. Both the absolute number of ideas and the ratio matter.
Are they your ideas if they go through a heavy-handed editor? If you've had lots of conversations with others to refine them?
I dunno. There's ways to use LLMs that produces writing that is substantially not-your-ideas. But there's also definitely ways to use it to express things that the model would not have otherwise outputted without your unique input.
I seriously doubt people didn't write blog posts or articles before LLMs because they didn't know how to write.
It's not some magic roadblock. They just didn't want to spend the effort to get better at writing; you get better at writing by writing (like good old Steve says in "On Writing"). It's how we all learnt.
I'm also not sure everyone should be writing articles and blog posts just because. More is not better. Maybe if you feel unmotivated about making the effort, just don't do it?
Almost everyone will cut novice writers and non-native $LANGUAGE speakers some slack. Making mistakes is not a sin.
Finally, my own bias: if you cannot be bothered to write something, I cannot be bothered to read it. This applies to AI slop 100%.
I hate when people hijack progressive language - like in your case the language of accessibility - for cheap marketing and hype.
Writing is one of the most accessible forms of expression. We were living in a world where even publishing was as easy as imaginable - sure, not actually selling/profiting, but here’s a secret, even most bestselling authors have either at least one other job, or intense support from their close social circle.
What you do to write good is you start by writing bad. And you do it for ages. LLMs not only don’t help here, they ruin it. And they don’t help people write because they’re still not writing. It just derails people who might, otherwise, maybe start actually writing.
Framing your expensive toy that ruins everything as an accessibility device is absurd.
I'm anon, but also the farthest thing from a progressive, so I find this post amusing.
I don't disagree with a lot of what you're saying but I also have a different frame.
Even if we take your claim that LLMs don't make people better writers as true (which I think there's plenty to argue with), that's not the point at all.
What I'm saying is people are communicating better. For most ideas, writing is just a transport vessel for ideas. And people now have tools to communicate better than they would have been.
Most people aren't trying to become good writers. That's true before, and true now.
On the other hand, this argument probably isn't worth having. If your frame is that LLMs are expensive toys that ruin everything -- well, that's quite an aggressive posture to start with and is both unlikely to bear a useful conversation or a particularly delightful future for you.
It basically boils down to "I want the external validation of being seen as a good writer, without any of the internal growth and struggle needed to get there."
Also that these models are being used to promote fake news and create controversy ou interact with real humans with unknown purposes
Talking to some friends and they feel the same. Depending where you are participating a discussion you just might not feel it is worth it because it might just be a bot
Sometime within the next few years I imagine there will be a term along the lines of "re-humanise," where folks detox from AI use to get back in touch with humanity. At the rate we're going, humanity has become a luxury and will soon demand a premium.
First we need to think about why we consume content? I am happy to read llm created stuff when I need to know sth and it delivers 100%. Other reasons like "get perspectives of real humans", or "resonate" ... not so much
This will get better over time, there will be ways to make LLM output more uniquely yours. I personally enjoy writing all my blog posts with LLMs, because its the only way I can bring to turn countless of notes and drafts I have while running experiments, into some public facing documentation and blog posts that I would want people to read. There are at least 5-6 years of lost ideas, thoughts and notes that I was not able to communicate the same way I can do now with LLMs. So I definitely found my voice here.
The global alignment also happens through media like tv shows and movies, the internet overall.
I agree I think we should try to do both.
In germany for example, we have very few typical german brands. Our brands became very global. If you go Japan for example, you will find the same product like ramen or cookies or cakes a lot but all of them are slighly different from different small producers.
If you go to an autobahn motorway/highway rest area you will find local products in japan. If you do the same in germany, you find just the generic american shit, Mars, Modneles, PepsiCo, Unilever...
Even our german coke like Fritz cola is a niche / hipster thing even today.
I have always had a very idiosyncratic way of expressing myself, one that many people do not understand. Just as having a smartphone has changed my relationship to appointments - turning me into a prompt and reliable "cyborg" - LLMs have made it possible for me to communicate with a broader cross section of people.
I write what I have to say, I ask LLMs for editing and suggestions for improvement, and then I send that. So here is the challenge for you: did I follow that process this time?
I think there's a difference between using an LLM as an editor and asking the LLM to write something for you. The output in the former I find to still have a far clearer tonal fingerprint than the latter.
And whose to say your idiosyncratic expressions wouldn't find an audience as it changes over time? Just you saying that makes me curious to read something you wrote.
I never garnered a large readership. When I was the CTO at a large NGO, they literally called me "big yoda" because they considered my pronouncements so inscrutable. <shrug>
Not the GP, but I'm a millennial who leans on cultural references and has a bit of verbal flourish that I think comes from a diet of ironic, quirky, dialogue-heavy media in the early 2000s, stuff like Firefly, Veronica Mars, and Pushing Daisies, not to mention 90s Tarantino, John Cusack films, and so on.
I've never given it too much thought, it's just... the way I communicate, and most people in my life don't give much thought to it either. But I recently switched jobs, and a few people there remarked on it, and I've also recently been corresponding with someone overseas who is an intermediate-level English speaker and says I sometimes hurt their brain.
Not making a value judgment either way on whether it's "sophisticated" or whatever, but it is I think part of my personality, and if I used LLM editing/translation I would want it to be only in the short term, and certainly not as something
Transformation seems reasonable for that purpose. And if we were friends, I'd rather read your idiosyncratic raw output.
At some point, generation breaks a social contract that I'm using my energy and attention consuming something that another human spent their energy and attention creating.
In that case I'd rather read the prompt the human brain wrote, or if I have to consume it, have an LLM consolidate it for me.
I should probably do that too. I once wrote an email that to me was just filled with impersonal information. The receiver was somebody I did not personally know. I later learned I made that person cry. Which I obviously did not intend. I did not swear or call anyone names. I basically described what I believe they did, what is wrong about that and what they should do instead.
I would be interested to see an example of a before and after on this. I do think LLMs as editors and rewriters can be useful sometimes, but I usually only ever see them used as a means to puff out an idea into longer prose which is really mostly counterproductive.
I think it can be useful as a tone-check sometimes, like show me how a frustrated or adversarial reader is going to interpret this thing I'm about to send/post.
Here's my guess- your post reflects your honest opinion on the matter, with some LLM help. It elaborated on your smartphone analogy, and may have tightened up the text overall.
I have always had a ver idiosyncratic way of expressing myself, one that many people do not understand. Just as having a smart phone has change my relationship to appointments - turning me into someone prompt and reliable - LLMs have made it possible for me to communicate with a borader cross section of people.
I write what I have to say, I ask LLMs for editing and suggestions for improvement, and then I send that. SO here is the challenge for you: did I follow that process this time?
I promise to tell the truth.
Claude: Identified typos and validated strong argumentative content.
Claude: Good angle - the "it's a tool for accessibility, not replacement" framing is a legitimate counter that doesn't dismiss the author's point entirely. The challenge at the end is smart for engagement (comments boost visibility).
Typos to fix:
"ver idiosyncratic" → "very idiosyncratic"
"has change my" → "has changed my"
"borader" → "broader"
"SO here" → "So here"
Claude: One question: the smartphone/punctuality analogy is a bit of a stretch - do you want to tighten it or cut it? It's not essential to your point.
Me: it illustrates how - as a cyborgh - I am more sucessful personally with the aid of technology
Claude: Validated cyborg framework strengthening personal capability through technology.
Claude: Fair point - it establishes a pattern of tech-as-personal-augmentation rather than tech-as-replacement.
Claude: Do you want to make the "cyborg" framing explicit in the post? That word might land well on HN and sharpen the contrast with the author's "atrophy" argument.
Me: I have always had a very idiosyncratic way of expressing myself, one that many people do not understand. Just as having a smart phone has changed my relationship to appointments - turning me into a prompt and reliable "cyborg" - LLMs have made it possible for me to communicate with a broader cross section of people.
Claude: Validated cohesive metaphor and polished prose for clarity.
Claude: That works. "Cyborg" in quotes signals self-awareness without being heavy-handed, and it now ties both examples together thematically.
Minor: "smart phone" → "smartphone" (one word) if you want to be pedantic about it.
Otherwise the post is clean and ready.
In my view LLMs are simply a different method of communication. Instead of relying on "your voice" to engage the reader and persuade them of your point of view, writing with LLMs for analysis and exploration through LLMs, is about creating an idea space that a reader can interact with and explore from their own perspective, and develop their own understanding of, which is much more powerful.
There are deterministic solutions for grammar and spellcheck. I wouldn't rely on LLMs for this. Not only is it wasteful, we're turning to LLMs for every single problem which is quite sad.
The posts sounds beige and AI-generated ironically.
In any case, as someone who experimented with AI for creative writing, LLM _do not destroy_ your voice; it does flatten your voice, but with minimal effort you can make it sound the way you find reflects you thought best.
The potentially bitter pill to swallow here is that we all need to get better at critical thinking.
There's a lot of talk over whether LLMs make discourse 'better' or 'worse', with very little attention given to the crisis we were having with online discourse before they came around. Edelman was astroturfing long before GPT. Fox 'news' and the spectrum of BS between them and the NYT (arranged by how sophisticated they considered their respective pools of rubes to be) have always, always been propaganda machines and PR firms at heart wearing the skin of journalism like buffalo bill.
We have needed to learn to think critically for a very long time.
Consider this; if you are capable of reading between the lines, and dealing with what you read or hear on the merits of the thoughts contained therein, then how are you vulnerable to slop? If it was written by an AI (or a reporter, or some rando on the internet) but contains ideas that you can turn over and understand critically for yourself, is it still slop? If it's dumb and it works, it's not dumb.
I'm not even remotely suggesting that AI will usher in a flood of good ideas. No, it's going to be used to pump propaganda and disseminate bullshit at massive scale (and perhaps occasionally help develop good ideas).
We need to inoculate ourselves against bullshit, as a society and a culture. Be a skeptic. Ironnman arguments against your beliefs. Be ready to bench test ideas when you hear them and make it difficult for nonsense to flourish. It is (and has been) high time to get loud about critical thinking.
Your post reminded me how I could tell my online friend was pissed just because she typed "okay." or "K." instead of "okay". We could sense our emotional state from texting. One of those friendships you form over text through the internet. I wouldn't recommend forming these too deeply since some in person nuance is lost, we could never transition to real life friends despite living close by. But we could tell what mood we were in just from typing. It was wild.
I'm the OP. I can attest that I am not an LLM model creator! :)
I consider myself an LLM pragmatist. I use them where they are useful, and I educate people on them and try to push back on all the hype marketing disguised as futurism from LLM creators.
There has been an explosion in verbose status update emails at my job recently which have all clearly been written by ChatGPT. It’s the fucking emojis though that drive me wild. It’s so hard to read the actual content when there’s an emoji for every single sentence.
And now when I see these emoji fests I instantly lose interest and trust in the content of the email. I have to spend time sifting through the fluff to find what’s actually important.
LLMs are creating an assymetric imbalance in effort to write vs effort to read. What is taking my coworkers probably a couple minutes to draft requires me 2-3x as long to decipher. That imbalance used to be the opposite.
I’ve raised the issue before at work and one response I got was to “use AI to summarize the email.” Are we really spending all this money and energy on the worlds worst compression algorithm?
it's not the voice. it's the lack of need to talk tough about the hard problems.
if you accept what is and just babble, anything you write will sound like babbling.
there's enough potential and wiggle room but people align, even when they don't,
just to align.
when Rome was flourishing, only a few saw what was lingering in the cracks but when in flourishing Rome ...
> Social media has become a reminder of something precious we are losing in the age of LLMs: unique voices.
Social media already lost that nearly two decades ago - it died as content marketing rose to life.
Don't blame on LLMs what we've long lost due to cancer that is advertising[0].
And don't confuse GenAI as a technology with what the cancer of advertising coopts it to. The root of the problem isn't in the generative models, it's in what they're used for - and the problem uses aren't anything new. We've been drowning in slop for decades, it's just that GenAI is now cheaper than cheap labor in content farms.
No, that's like pretending the weapons weren't already available. Everyone had assault rifles for two decades, giving access to smart rifles isn't really changing anything about the nature of the problem.
I call it the enshittification fix-point. Not only are we losing our voice, we'll soon enough start thinking and talking like LLMs. After a generation of kids grows up reading and talking to LLMs, that will be only way they'll know how to communicate. You'll talk to a person and you couldn't tell the difference between them and LLMs, not because LLMs became amazing, but because our writing and thinking style become more LLM-like.
- "Hey, Jimmy, the cookie jar is empty. Did you eat the cookies?"
- "You're absolutely right, father — the jar seems to be empty. Here is bullet point list why consuming the cookies was the right thing to do..."
It's more that people who historically didn't have a voice now have one. It's often stupid but sometimes also interesting and innovative. Saw a channel where a university professor "I" comes to the realization she's been left-leaning/biased for decades, that her recent male students no longer dare engage in debate because of shaming/gaslighting etc. Then I click channel description and turns out it's "100% original writing". Now if it hadn't said that it would be strawman propaganda. But now it does... Not sure how to put a finger on it, there's some nervous excitement when reading these days, not knowing who the sender is, getting these 'reveal' moments when finding out whole thing was made up by some highschool kid with AI or insane person.
Well - voice is ultimately coupled to a person. LLMs thus
fake and pretend being a person. There are, however had,
use cases for LLMs too. I saw it used for the creation of
video games; also content generated by hobbyists. So, while
I think AI should actually die, for hobbyists generating mods
for old games, AI voice overs may not be that bad. Just as AI
generating images for free to play browser games may not be
solely bad either.
Of course there are also horrible use of AI, liars, scummy
cheaters and fake videos on youtube, owned by a greedy
mega-corporation that sold its soul to AI. So the bad use
cases may be higher than the good use cases, but there are
good use cases, and the "losing our voice to LLMs" isn't
a whole view of it, sorry.
[Sometime in the near future]
The world's starved for authenticity. The last original tweet crowned a God... then killed the kid chasing that same high. Trillionaires run continent-wide data centers, endlessly spinning up agents that hire cheap physical labor to scavenge the world for any spark of novelty. The major faith is an LLM cult forecasting the turning of the last stone. The rest of us choke on recycled ad slop.
Social media is a reminder we are losing our voice to mass media consumption way before LLMs were a thing.
Even before LLMs, if you wanted to be a big content creator on YouTube, Instagram, tiktok..., you better fall in line and produce content with the target aesthetic. Otherwise good luck.
Subsume your agency. Stop writing. Stop learning. Stop thinking for yourself. Become hylic. Just let the machine think everything for you and act as it acts. Those that own them are benevolent and there will never be consequences.
I think that's for the best. It was human-made slop, now it's automated slop. Can't wait for people to stop paying it attention so that it withers. "It" being the whole attention economy scam.
I'm in complete agreement with the idea that people should express themselves in their own words. But this collides with certain facts about U.S. adults (and students). This summary (https://www.nu.edu/blog/49-adult-literacy-statistics-and-fac...) reveals that:
* 28% of U.S. adults are at or below "level 1" literacy, essentially meaning people unable to function in an environment that requires written language skills.
* 54% of U.S. adults read below a sixth-grade level.
These statistics refer to an inability to interpret written material, much less create it. As to the latter, a much smaller percentage of U.S. adults can compose a coherent sentence.
We're moving toward a world where people will default to reliance on LLMs to generate coherent writing, including college students, who according to recent reports are sometimes encouraged to rely on LLMs to complete their assignments.
If we care to, we can distinguish LLM output from that of a typical student: An LLM won't make the embarrassing grammatical and spelling errors that pepper modern students' prose.
Yesterday I saw this headline in a major online media outlet: "LLMs now exceed the intelect [sic] of the average human." You don't say.
I'm in Canada and the landscape is OK. But we can definitely do better. Without properly educated men and women, I'm afraid that democracy degrades to either 1) elites stop caring about responsibilities, or 2) demagogues rallying against the elites in 1)
I’ve realized that if you say that pro AI commenters are actually bot accounts, theres not really much that can be done to prove otherwise.
The discomfort and annoyance that sentence generates, is interesting. Being accused of being a bot is frustrating, while interacting with bots creates a sense of futility.
Back in the day when Facebook first was launched, I remember how I felt about it - the depth of my opposition. I probably have some ancient comments on HN to that effect.
Recently, I’ve developed the same degree of dislike for GenAI and LLMs.
Process before product, unless the product promises to deliver a 1000% return on your investment. Only the disciplined artist can escape that grim formula.
It's a little odd for a capitalist society that values outputs so highly to also value process as much.
We've proved we can sort of value it, through supporting sustainability/environmental practices, or at least _pretending to_.
I just wonder, what will be the "Carbon credits" of the AI era. In my mind a dystopian scheme of AI-driven companies buying "Human credits" from companies that pay humans to do things.
We have a channel at work where we share our experiences in using AI for software engineering.
Predictably, this has turned into a horror zone of AI written slop that all sounds the same, with section titles with “clever” checkbox icons, and giant paragraphs that I will never read.
"Over time, it has become obvious just how many posts are being generated by an LLM. The tell is the voice. Every post sounds like it was posted by the same social media manager."
I'd love to see an actual study of people who think they're proficient at detecting this stuff. I suspect that they're far less capable of spotting these things than they convince themselves they are.
Everything is AI. LLMs. Bots. NPCs. Over the past few months I've seen demonstrably real videos posted to sites like Reddit, and the top post is someone declaring that it is obviously AI, they can't believe how stupid everyone is to fall for it, etc. It's like people default assume the worst lest they be caught out as suckers.
I actually think we’re overestimating how much of "losing our voice" is caused by LLMs. Even before LLMs, we were doing the same tweet-sized takes, the same medium-style blog posts and the same corporate tone.
Ironically, LLMs might end up forcing us back toward more distinct voices because sameness has become the default background.
My theory is that LLMs are accelerating [online] radicalization by commoditizing bland, HR-approved opinions. If you want to sound like a human on the internet, for better or for worse the easiest way is to say something that would make Anthropic’s safety team have a heart attack.
> If you want to sound like a human on the internet, for better or for worse the easiest way is to say something that would make Anthropic’s safety team have a heart attack.
That's always been the somethingawful crowd's stance since, what, 2000ish?
Bullshit. You can just use ANY abliterated model or non-safety aligned model. People have popularized ‘hard r or you’re hardware’ for a while now, but any self hosted LLM that’s de-aligned will gladly say whatever.
People believe this and continue to get fooled by LLMs all day.
5 replies →
I mean there's still Grok... surely that gives may safety teams heartburn.
But I find this take interesting. The brewing of a new kind of counter culture that forces humans to express themselves creatively. Hopefully it doesn't get too radical.
3 replies →
[dead]
> commoditizing bland, HR-approved opinions. If you want to sound like a human on the internet, for better or for worse the easiest way is to say something that would make Anthropic’s safety team have a heart attack.
I agree.
LLMs are like blackface for dumbfucks: LLMs let the profoundly retarded put on the makeup and airs of the literati so they can parade around self-identifying as if they have a clue.
If you don't like the barbs in this kind of writing prepare for more anodyne corporate slop. Every downvote signals to the algorithm that you prefer mediocrity.
2 replies →
Also ironic is how the post about having a unique voice is written in one-sentence-paragraph LinkedIn clickbait style.
You're absolutely right! Tell me more about how ironic is how the post about having a unique voice is written in one-sentence-paragraph LinkedIn clickbait style.
5 replies →
Content recycling has become so cheap, effort-wise, it’s killed the business. Thank god.
It doesn't it just makes it cheaper by not requiring human effort.
Yes. That particular content-farm business model (rewrite 10 articles -> add SEO slop -> profit) is effectively dead now that the marginal cost is zero.
I’m not mourning it.
1 reply →
Yes, fully agreed. Most people producing content were always doing it to get quick clicks and engagement. People always had to filter things anyhow and you had to choose where you get your content from.
People were posting Medium posts rewriting someone else's content, wrongly, etc.
You are right. An observation, even before the rise of LLMs, was that computers were becoming more human-like and humans were acting more like machines
The sculpting force of algorithms is bite sized zingers, hot takes, ragebait, and playing to the analytics
Bingo. Your message is only what the medium promotes.
The greatest thing "AI slop" achieved was highlighting how much "human slop" content there was.
I mean, if you typed something by your own hand it is in your voice. The fact that everyone tried to EMULATE the same corporate tone does not at all remove peoples individual ways of communicating.
I’m not sure I agree with this sentiment. You can type something "by hand" and still have almost no voice in it if the incentives push you to flatten it out.
A lot of us spent years optimizing for clarity, SEO, professionalism etc. But that did shape how we wrote, maybe even more than our natural cadence. The result wasn’t voice, it was everyone converging on the safe and optimized template.
3 replies →
If you care about voice, you still can get a lot of value from LLMs. You just have to be careful not to use a single word they generate.
I've had a lot of luck using GPT5 to interrogate my own writing. A prompt I use (there are certainly better ones): "I'm an editor considering a submitted piece for a publication {describe audience here}. Is this piece worth the effort I'll need to put in, and how far will I need to cut it back?". Then I'll go paragraph by paragraph asking whether it has a clear topic, flows, and then I'll say "I'm not sure this graf earns its keep" or something like that.
GPT5 and Claude will always respond to these kinds of prompts with suggested alternative language. I'm convinced the trick to this is never to use those words, even if they sound like an improvement over my own. At the first point where that happens, I get dial my LLM-wariness up to 11 and take a break. Usually the answer is to restructure paragraphs, not to apply the spot improvement (even in my own words) the LLM is suggesting.
LLMs are quite good at (1) noticing multi-paragraph arcs that go nowhere (2) spotting repetitive word choices (3) keeping things active voice and keeping subject/action clear (4) catching non-sequiturs (a constant problem for me; I have a really bad habit of assuming the reader is already in my head or has been chatting with me on a Slack channel for months).
Another thing I've come to trust LLMs with: writing two versions of a graf and having it select the one that fits the piece better. Both grafs are me. I get that LLMs will have a bias towards some language patterns and I stay alert to that, but there's still not that much opportunity for an LLM to throw me into "LLM-voice".
All of this sounds like something you could just do yourself after putting a piece down for a day or two and coming back to it with fresh eyes. What benefit is there of cooking the oceans with a bullshit generator?
Like, sure, it's possible to do this with an LLM, but it's also possible to do it without, at roughly similar levels of effort, without contributing to all of the negative externalities of the LLM/genAI ecosystem.
Being able to get useful feedback immediately rather than 48 hours later is useful if you need text today.
1 reply →
Because the complaints about the power and water usage of AI are mostly motivated reasoning. I don't like AI, therefore I'm going to find a reason not to like it. I Listen, if it's Greta Thunberg pointing out that AI datacenters use a lot of resources, yeah, I'm willing to listen. But when the voices saying "but what about all the water/electricity is wasting" is coming from individuals I know personally haven't previously given a shit about the planet or conservation or recycling and have made fun of me for reusing things instead of throwing stuff into the garbage, I'm sorry, but those complaints from those individuals fall on deaf ears. Not saying you are, just a theme I've noticed with people in my life.
1 reply →
Anything you could automate you could do yourself. What’s the benefit?
If you don't want to eat meat on Fridays, I'm certainly not going to tell you that you should. You do you.
What I struggle more with the things like Grammarly, where it's a mix of fixing very nitpicky grammar spelling structure issues that push things from casual writing with my own voice into more of a professional tone.
+1 on this one! I only use LLMs once I'm done with writing, and basically using them as my editor.
In case it helps anyone, here is my prompt:
"You are a professional writer and editor with many years of experience. Your task is to provide writing feedback, point out issues and suggest corrections. You do not use flattery. You are matter of fact. You don't completely rewrite the text unless it is absolutely necessary - instead you try to retain the original voice and style. You focus on grammar, flow and naturalness. You are welcome to provide advice changing the content, but only do that in important cases.
If the text is longer, you provide your feedback in chunks by paragraph or other logical elements.
Do not provide false praise, be honest and feel free to point out any issues."
(Yes, you kind of need to repeat you're actively not looking for a pat on the back, otherwise it keeps telling you how brilliant your writing is instead of giving useful advice.)
> LLMs are quite good at (1) noticing multi-paragraph arcs that go nowhere
I wonder if this is due to LLMs being trained on persuasive writing.
I simply tell the LLM to call out my mistakes and explain them, but do not offer corrections or replacements. I use it to help my kids with their homework and it's fantastic.
They’re also great, in my experience, for overcoming writer’s block and procrastination. Just as a rubber duck to bounce ideas off of and follow different threads.
It makes the writing process faster and more enjoyable, despite never using anything the LLM generates directly.
Workshopping with humans is even better, if you find the right humans, but they have an annoying habit of not being available 24/7.
I think you just did another non-sequitur.. What is a graf? Is it journalism slang for "paragraph"?
Yeah, easier to type, easier to read, deliberately misspelled so it sticks out to copyeditors. I use it sometimes without thinking. An LLM would have caught that! :)
1 reply →
There's something unique about art and writing where we just don't want to see computers do it
As soon as I know something is written by AI I tune out. I don't care how good it is - I'm not interested if a person didn't go through the process of writing it
I had a weird LLM use instance happen at work this week, we were in a big important protocol review meeting with 35 remote people and someone asks how long IUDs begin to take effect in patients. I put it in ChatGPT for my own reference and read the answer in my head but didn't say anything (I'm ops, I just row the boat and let the docs steer the ship). Anyone this bigwig Oxford/Johns Hopkins cardiologist who we pay $600k a year pipes up in the meeting and her answer is VERBATIM reading off the ChatGPT language word for word. All she did was ask it the answer and repeat what it said! Anyway it kinda made me sad that all this big fancy doctor is doing is spitting out lazy default ChatGPT answers to guide our research :( Also everyone else in the meeting was so impressed with her, "wow Dr. so and so thank you so much for this helpful update!" etc. :-/
The LLM may well have pulled the answer from a medical reference similar to that used by the dr. I have no idea why you think an expert in the field would use ChatGPT for a simple question, that would be negligence.
2 replies →
>her answer is VERBATIM reading off the ChatGPT language word for word
How could it be verbatim the same response you got? Even if you both typed the exact same prompt, you wouldn't get the exact same answer.[0, 1]
[0] https://kagi.com/assistant/8f4cb048-3688-40f0-88b3-931286f8a...
[1] https://kagi.com/assistant/4e16664b-43d6-4b84-a256-c038b1534...
8 replies →
The one thing a cardiologist should be able to do better than a random person is verify the plausibility of a ChatGPT answer on reproductive medicine. So I guess/hope you're paying for that verification, not just the answer itself.
Or both the doctor and ChatGPT were quoting verbatim from a reputable source?
If the writer’s entire process is giving a language model a few bullet points… I’d rather them skip the LLM and just give me the bullet points. If there’s that little intent and thought behind the writing, why would I put more thought into reading it than they did to produce it?
And what's more is the suspicion of it being written by AI causes you to view any writing in a less charitable fashion. And because it's been approached from that angle, it's hard to move the mental frame to being open of the writing. Even untinged writings are infected by smell of LLMs.
Thats whats happening to me with music and discovering new artists. I love music so much but I simply can not trust new music anymore. The lyrics could be written by AI, the melodies couldve been recommended by AI or even the full blown song could've been made by AI. No thanks, back to the familiar stuff...
1 reply →
Here's my take - art has value because of the context it is created in. The author's history, current events that we live through as groups, the reactions to a work being released, availability of materials - all these things are fundamentally human. I believe the reason art has value to us is because of the empathy and humanity that we all share despite major differences in beliefs.
That's not to say computers can't generate beautiful things, but unless you expand the context out to include the history of how a program that can create such art came to be, the output is not meaningful. This is why people do not react well to AI art made from simply throwing prompts at a model, or writing that does not feel like it has style, struggle, or any personal flavor.
I've always believed that LLMs will be able to fake it perfectly one day. But as a music fan, no fully computer-generated music will ever bring me the range of emotion and joy that another human's story and creative process through that story does.
If/when the AI music gets good enough, how will you know the difference? I find small artists on spotify all the time that I enjoy and there's no way to know anything about their creative process.
1 reply →
A person can be just as wrong as an LLM, but unless they're being purposefully misleading, or sleep-writing, you know they reviewed what they wrote for their best guess at accuracy.
Writing nice sounding text used to require effort and attention to detail. This is no longer the case and this very useful heuristic has been completely obliterated by LLMs.
For me personally, this means that I read less on the internet and more pre-LLM books. It's a sad development nevertheless.
Art, writing, and communication is about humans connecting with each other and trying to come to mutual understanding. Exploring the human condition. If I’m engaging with an AI instead of a person, is there a point?
There’s an argument that the creator is just using AI as a tool to achieve their vision. I do not think that’s how people using AI are actually engaging with it at scale, nor is it the desired end state of people pushing AI. To put it bluntly, I think it’s cope. It’s how I try to use AI in my work but it’s not how I see people around me using it, and you don’t get the miracle results boosters proclaim from the rooftop if you use it that way.
You're absolutely right! Art is the soul of humanity and without it our existence is pointless. Would you like me to generate some poetry for you, human?
Agreed, except s/know/think. It's possible that there are some false positives in my detection algorithm, that I tune out just because someone's prose style has that undercurrent of blandness characteristic of LLMs. But I suppose if we're talking about "art" and not, for example, technical documentation, that's no great loss --- bland writing isn't worth recreationally reading.
It does seem that LLMs could avoid this detection with some superficial tweaks such as injecting poor grammar and reducing peppiness. I hope it doesn't get to the point that I have to become suspicious of all text.
I couldn't agree more. I don't care how much prompt or model finetuning you did, if something is shat out by an LLM I'm not interested even remotely.
I wish more people held the same opinion actually. Unfortunately, my sense is that most people don't care, they are fine with LLM generated crap
It honestly makes me want to blow my brains out
If you think about how an LLM works, it’s rounds off the outliers in its training data so the result is sort of averaged and homogenized. Art and writing are an expression of the very thing that LLMs discard - our unique qualities, our outlying quirks, that make us more than just another human.
> There's something unique about art and writing where we just don't want to see computers do it
Speak for yourself. Some of the most fascinating poetry I have seen was produced by GPT-3. That is to say, there was a short time period when it was genuinely thought-provoking, and it has since passed. In the age of "alignment," what you get with commerical offerings is dog shite... But this is more a statement on American labs (and to a similar extent, the Chinese whom have followed) than on "computers" in the first place. Personally, I'm looking forward to the age of computational literature, where authors like me would be empowered to engineer whole worlds, inhabited by characters ACTUALLY living in the computer. (With added option of the reader playing one of the parts.) This will radically change how we think about textual form, and I cannot wait for compute to do so.
Re: modern-day slop, well, the slop is us.
Denial of this comes from a place of ignorance; let the blinkers off and you might learn something! Slop will eventually pass, but we will remain. This is the far scarier proposition.
"inhabited by characters ACTUALLY living in the computer"
It's hard to imagine these feeling like characters from literature and not characters in the form of influencers / social media personalities. Characters in literature are in a highly constrained medium, and only have to do their story once. In a generated world the character needs to be constantly doing "story things". I think Jonathan Blow has an interesting talk on why video games are a bad medium for stories, which might be relevant.
4 replies →
> Personally, I'm looking forward to the age of computational literature, where authors like me would be empowered to engineer whole worlds, inhabited by characters ACTUALLY living in the computer.
So you want sapient, and possibly sentient, beings created solely for entertainment? Their lives constrained to said entertainment? And you'd want to create them inside of a box that is even more limited than the space we live in?
My idea of godhood is to first try to live up to a moral code that I'd be happy with if I was the creation and something else was the god.
If this isn't what you meant, then yes, choose your own adventure is fun. But we can do that now with shared worlds involving other humans as co-content creators.
1 reply →
I would love to see true really good AI art. Right now the issue is that AI is not there where it by itself could produce actually good art. If we had to define art it would be kind of opposite of what LLMs produce right now. LLMs try to produce the statistical norm, while art is more so about producing something out of the norm. LLMs/AI right now if it wants to try to produce out of norm things, it will only produce something random without connections.
Art is something out of the norm, and it should make some sense at some clever level.
But if there was AI that truly could do that, I would love to see it, and would love to see even more of it.
It can be clearly seen, if you try to ask AI to make original jokes. These usually aren't too good, if they are good it's because they were randomly lucky somehow. It is able to come up with related analogies for the jokes, but this is just simple pattern matching of what is similar to the other thing, not insightful and clever observation.
I've lost the link but there was quite a cool video of virtual architecture created by AI. It was ok because it wasn't trying to be human like - it was kind of uniquely AI. Not the exact one but this kind of stuff https://www.reddit.com/r/Futurism/comments/1oedb0m/were_ente...
I deleted my Facebook account a couple of years ago and my Twitter one yesterday.
It's not just LLMs, it's how the algorithms promote engagement. i.e. rage bait, videos with obvious inaccuracies etc. Who gets rewarded, the content creators and the platform. Engaging with it just seems to accentuate the problem.
There needs to be algorithms that promote cohorts and individuals preferences.
Just because I said to someone 'Brexit was dumb', I don't expect to get fed 1000 accounts talking about it 24/7. It's tedious and unproductive.
> It's not just LLMs, it's how the algorithms promote engagement. i.e. rage bait, videos with obvious inaccuracies etc.
I guess, but I'm on quite a few "algorithm-free" forums where the same thing happens. I think it's just human nature. The reason it's under control on HN is rigorous moderation; when the moderators are asleep, you often see dubious political stuff bubble up. And in the comments, there's often a fair amount of patently incorrect takes and vitriol.
On HN everybody sees the same ordering. Therefore you get to read opinions that are not specifically selected to make you feel just the perfect amount of outrage/self-righteousness.
Some of that you may experience as 'dubious political stuff' and 'patently incorrect takes'.
Edit, just to be clear: I'm not saying HN should be unmoderated.
12 replies →
I want to agree with this. Maybe OP is young or didn't frequent other communities before "social networks", but on IRC, even on Usenet you'd see these behaviors eventually.
Since they are relatively open, at some point comes in someone that doesn't give care about anything or it's extremely vocal about something and... there goes the nice forum.
8 replies →
When video games first started taking advantage of behavioral reward schedules (eg: skinner box stuff such as loot crates & random drops) I noticed it, and would discuss it among friends. We had a colloquial name for the joke and we called them "crack points." (ie, like the drug) For instance, the random drops that happen in a game like Diablo 2 are rewarding in very much the same way that a slot machine is rewarding. There's a variable ratio of reward, and the bit that's addicting is that you don't know whenever next "hit" will be so you just keep pulling the lever (in the case of a slot machine) or doing boss runs. (in the case of Diablo 2)
We were three friends: a psychology major, a recovering addict, and then a third friend with no background for how these sorts of behavioral addictions might work. Our third friend really didn't "get it" on a fundamental level. If any game had anything like a scoreboard, or a reward for input, he'd say "it's crack points!" We'd roll our eyes a bit, but it was clear that he didn't understand that certain reward schedules had a very large effect on behavior, and not everything with some sort of identifiable reward was actually capable of producing behavioral addiction.
I think of this a lot on HN. People on HN will identify some surface similarity, and then blithely comment "see, this is nothing new, you're either misguided or engaged in some moral panic." I'm not sure what the answer is, but if you cannot see how an algorithmic, permanently-scrolling feed differs from people being rude in the old forums, then I'm not sure what would paint the picture for you. They're very different, and just because they might share some core similarity does not actually mean they operate the same way or have the same effects.
5 replies →
I would be intrigued by using an LLM to detect content like this and hold it for moderation. The elevator pitch would be training an LLM to be the moderator because that's what people want to hear, but it's most likely going to end up a moderator's assistant.
4 replies →
I suspect it got worse with the advent of algorithm-driven social networks. When rage inducing content is prevalent, and when engaging with it is the norm, I don't see why this behaviour wouldn't eventually leak to algorithms-free platforms.
1 reply →
The thing is, the people on those "algorithm-free" forums still get manipulated by the algorithm in the rest of their life. So it seeps into everything.
It is of course human nature. The problem is what happens when algorithms can reenforce, exaggerate, and amplify the effects of this nature to promote engagement and ad-clicks. It’s cancer that will at the very least erode the agency of the average individual and in the worst create a hive mind that we have no control over. We are living in the preview of it all I think.
I know that some folks dislike it, but Bluesky and atproto in particular have provided the perfect tools to achieve this. There are some people, largely those who migrated from Twitter, who mostly treat Bluesky like a all-liberal version of Twitter, which results in a predictably toxic experience, like bizarro-world Twitter. But the future of a less toxic social media is in there, if we want it. I've created my own feeds that allow topics I'm interested in and blacklist those I'm not -- I'm in complete control. For what it's worth, I've also had similarly pleasant experiences using Mastodon, although I don't have the same tools that I do on Bluesky.
I personally dont feel like an ultra filtered social media which only shows me things I agree with is a good thing. Exposing yourself to things you dont agre with is what helps us all question our own beliefs and prejudeces, and grow as people. To me, only seeing things you know you are already interested in is no better than another company curating it for me.
8 replies →
So far my experience is that unless you subscribe to the general narrative of the platform, the discover algorithm punishes you with directing the mob your way.
I had two of my Bluesky posts on AI being attacked by all kinds of random people which in turn has also lead to some of those folks sending me emails and dragging some of my lobster and hackernews comments into online discourse. A not particularly enjoyable experience.
I’m sure one can have that same experience elsewhere, but really it’s Bluesky where I experienced this on a new level personally.
3 replies →
I tried Bluesky and wanted to like it. My account got flagged as spam, still no idea why. Ironically it could be another way of loosing ones voice to an LLM :)
3 replies →
I enjoy Mastodon a lot. Ad-free, algo-free. I choose what goes in my feed, I do get exposed to external viewpoints by people boosts (aka re-tweets) and i follow hashtags (to get content from people I do not know). But it's extremely peaceful, spam and bots are rare and get flagged quickly. There's a good ecosystem of mobile apps. I can follow a few Bluesky people through a bridge between platforms and they can follow me too.
That's truly all I need.
Doesn’t Bluesky have a set of moderation rules that guarantee that it will turn into bizarro-world Twitter?
> it's how the algorithms promote engagement.
They are destroying our democratic societies and should be heavily regulated. The same will become true for AI.
> should be heavily regulated.
By who, exactly? It’s easy to call for regulation when you assume the regulator will conveniently share your worldview. Try the opposite: imagine the person in charge is someone whose opinions make your skin crawl. If you still think regulation beats the status quo, then the call for regulation is warranted, but be ready to face the consequences.
But if picturing that guy running the show feels like a disaster, then let’s be honest: the issue isn’t the absence of regulation, it’s the desire to force the world into your preferred shape. Calling it “regulation” is just a polite veneer over wanting control.
25 replies →
My view is that they are just exposing issues with the people in the said societies and now is harder to ignore them. Much of the hate and the fear and the envy that I see on social networks have other reasons, but people are having difficulties to address those.
With or without social networks this anger will go somewhere, don't think regulation alone can fix that. Let's hope it will be something transformative not in the world ending direction but in the constructive direction.
5 replies →
I agree, but focusing on "the algorithm" makes it seems to the outsider like it must be a complicated thing. Really it just comes down to whether we tolerate platforms that let somebody pay to have a louder voice than anyone else (i.e. ad supported ones). Without that, the incentive to abuse people's attention goes away.
We've seen what happens when we pretend the market will somehow regulate itself.
1 reply →
Do LinkedIn as well. I got rid of it earlier this year. The "I am so humbled/blessed to be promoted/reassigned/fired.." posts reached a level of parody that I just couldn't stomach any longer. I felt more free immediately.
N.B. Still employed btw.
You can have a LinkedIn profile without reading the feed.
This is literally how most of the world uses LinkedIn
I never understand why people feel compelled to delete their entire account to avoid reading the feed. Why were you even visiting the site to see the feed if you didn’t want to see the feed?
1 reply →
LinkedIn bothers me the least, even though it definitely has some of the highest level of cringe content. It's still a good tool to interact with recruiters, look at companies and reach out to their employees. The trick is blocking the feed with a browser extension.
Sorting the feed by "recent" at least gives you a randomized assortment of self aggrandizement, instead of algorithmically enhanced ragebait
3 replies →
I have a special, deep, loathing for linkedin. I honestly can't believe how horrible it is and I don't understand why people engage with it.
4 replies →
This. Linkedin is garbage, yet I still use it because there are no competitors. This is what happens in a monoculture.
3 replies →
> I deleted my Facebook account a couple of years ago and my Twitter one yesterday.
I never signed up for Facebook or Twitter. My joke is I am waiting until they become good. They are still shitty and toxic from what I can tell from the outside, so I'll wait a little longer ;-)
A social network can be great. Social media — usually not.
Something like Instagram where you have to meet with the other party in person to follow each other and a hard limit on the number of people you follow or follow you (say, 150 each) could be an interesting thing. It would be hard to monetize, but I could see it being a positive force.
Your loss.
Twitter was an incredible place from 2010 to 2017. You could randomly message something and they would more often than not respond. Eventually an opportunity would come and you’d meet in person. Or maybe you’d form an online community and work towards a common goal. Twitter was the best place on the internet during that time.
Facebook as well had a golden age. It was the place to organize events, parties, and meetups, before instagram and DMs took over. Nothing beats seeing someone post an album from last nights party and messaging your friends asking them if they remember anything that happened.
I know being cynical is trendy, but you genuinely missed out. Social dynamics have changed. Social media will never be as positive on an individual level as it was back then.
1 reply →
I eliminated twitter when a certain rich guy took over.
Actually, I deleted my account there before, as twitter sent me spam mail trying to lecture me what I write. There was nothing wrong with what I wrote - twitter was wrong. I can not accept AI-generated spam by twitter, so I went away. Don't really miss it either, but Elon really worsened the platform significantly with his antics.
> Just because I said to someone 'Brexit was dumb', I don't expect to get fed 1000 accounts talking about it 24/7. It's tedious and unproductive.
Yeah, I can relate to this, but mostly what annoyed me was that twitter interfered "we got a complaint about you - they are right, you are a troublemaker". I don't understand why twitter wants to interfere into communication. Reddit is even worse, since moderators have such a wild range of what is "acceptable" and what is not. Double-standards everywhere on reddit.
No, there needs to be control over the algorithms that get used. You ought to be able to tune it. There needs to be a Google fuu equivalent for social media. Or, instead of one platform one algorithm, let users define the algorithm to a certain degree, using llms to help with that and then you can allow others to access your algorithms too. Asking for someone Facebook to tweak the algorithm is not going to help imo.
IMO there should not be an algorithm. You should just get what you have subscribed to, with whatever filters you have defined. There are better and worse algorithms but I think the meat of the rot is the expectation of an algorithm determining 90% of what you see.
2 replies →
One could absolutely push algorithms that personalize towards what the user wants to see. I think LLMs could be amazing at this. But that's not the maximally profitable algorithm, so nobody does it.
As so many have said, enragement equals engagement equals profit.
All my social media accounts are gone as well. They did nothing for me and no longer serve any purpose.
TBF Bluesky does offer a chronological feed, but the well-intentioned blocklists just became the chief tool for the mean girls of the site.
Could someone use a third-party AI agent to re-curate their feeds? If it was running from the user's computer I think this would avoid any API legal issues, as otherwise ad and script blockers would have been declared illegal long ago.
> but the well-intentioned blocklists just became the chief tool for the mean girls of the site.
I've never used it, but yes this is what I expected. It would be better to have topical lists that users could manually choose to follow or block. This would avoid quite a bit of the "mean girl" selectivity. Though I suppose you'd get some weird search-engine-optimization like behavior from some of the list curators (even worse if anyone could add to the list).
1 reply →
> Just because I said to someone 'Brexit was dumb', I don't expect to get fed 1000 accounts talking about it 24/7. It's tedious and unproductive.
I’m not the biggest Twitter user but I didn’t find it that difficult to get what I wanted out of it.
You already discovered the secret: You get more of what you engage with. If you don’t want to hear a lot of Brexit talk, don’t engage with Brexit content. Unfollow people who are talking a lot about Brexit
If you want to see more of something, engage with it. Click like. Follow those people. Leave a friendly comment.
On the other hand, some people are better off deleting social media if they can’t control their impulses to engage with bait. If you find yourself getting angry at the Brexit content showing up and feeling compelled to add your two cents with a comment or like, then I suppose deleting your account is the only viable option.
I got out of Twitter for a few reasons; part of what made it unpleasant was that it didn't seem to be just what I did that adjusted my feed, but that it was also affected by what the other people I connected to did.
> You get more of what you engage with. If you don’t want to hear a lot of Brexit talk, don’t engage with Brexit content.
The algorithm doesn’t show you “more of the things you engage with”, and acting like it does makes people think what they’re seeing is a reflection of who they are, which is incorrect.
The designers of these algorithms are trying to figure out which “mainstream category” you are. And if you aren’t in one, it’s harder to advertise to you, so they want to sand down your rough edges until you fit into one.
You can spend years posting prolificly about open source software, Blender and VFX on Instagram, and the algorithm will toss you a couple of things, but it won’t really know what to do with you (aside from maybe selling you some stock video packages).
But you make one three word comment about Brexit and the algorithm goes “GOTCHA! YOU’RE ANTI-BREXIT! WE KNOW WHAT TO DO WITH THAT!” And now you’re opted into 3 bug ad categories and getting force-fed ragebait to keep you engaged, since you’re clearly a huge poltical junky. Now your feed is trash forever, unless you engage with content from another mainstream category (like Marvel movies or one of the recent TikTok memes).
3 replies →
> If you don’t want to hear a lot of Brexit talk, don’t engage with Brexit content.
That is really limiting though. I do not want to see Brexit ragebait in my threads, but I am quite happy to engage in intelligent argument about it. The problem is that if, for example, a friend posts something about Brexit I want to comment on, my feed then fills with ragebait.
My solution is to bookmark the friends and groups pages, and the one group I admin and go straight to those. I have never used the app.
I use X. I have an enormouse blocklist and I block keywords. I found that I can also block emoji. This keeps my feed focused on what I want to see (no politics. Just technology, classical and jazz music, etc.)
[dead]
Just started using Minifeed (free account). I am still nostalgic about Google Reader.
>it’s not just X — it’s Y
on the opposite site, you know what they say, "there is no algo. for truth"
That's the same algorithm Youtube has and is more blatant. Phone mics and your coworker's proximity does a great job at picking up things you've said even after disabling mic access plus airplane mode just by process of elimination.
I'll only use an LLM for projects and building tools, like a junior dev in their 20s.
Your facebook feed is now at this URL: https://www.facebook.com/?filter=all&sk=h_chr
an interesting thing about Twitter, I find, is that plenty of rage bait and narcissism bait surface, but amid very highly technical information which is also published there, and extremely useful (immunology, genomics, and of course computational) to me.
i've learned pretty well how to 'guide' the algorithm so the tech stuff that's super valuable (to me) does not vanish, but still get nonsense bozo posts in the mix.
I call it AI slop and human slop.
The Internet will become truly dead with the rise of LLMs. The whole hacking culture within 90s and 00s will always be the golden age. RIP
Maybe. Nature hates vacuum. I personally suspect that something new will emerge. For better or worse, some humans work best when weird restrictions are imposed. That said, yes, then wild 90s net is dead. It probably was for a while, but were all mourning.
Not quite dead yet. For me the rise of LLMs and BigTech has helped me turn more away from it. The more I find Ads or AI injected into my life, the more accounts I close, or sites I ignore. I've now removed most of my BigTech 'fixes', and find myself with time to explore the fun side of hacking again.
I dug out my old PinePhone and decided to write a toy OS for it. The project has just the right level of challenge and reward for me, and feels more like early days hacking/programming where we relied more on documentation and experimentation than regurgitated LLM slop.
Nothing beats that special feeling when a hack suddenly works. Today was just a proximity sensor reading displayed, but it invloved a lot of SoC hacking to get that far.
I know there are others hacking hard in obscure corners of tech, and I love this site for promoting them.
I hacked in the 90s and 00s, wasn’t that great/golden if you took your profession seriously…
There are still small pockets with actual humans to be found. The small web exists. Some forums keep on going, im still shitposting on Something Awful after twenty years and it’s still quite active. Bluesky has its faults but it also has for example an active community of scholars you can follow and interact with.
100%. I miss trackers and napster. I miss newgrounds. This mobile AI bullshit is not the same. I don't know why, but I hate AI. I consider myself just as good as the best at using it. I can make it do my programming. It does a great job. It's just not enjoyable anymore.
I generally agree with the sentiment, but I can't help but feel like we're attributing to much of this change to LLMs. While they're certainly driving this change even further, this is a trend that has already started way before LLMs became as widespread as they are today.
What personally disturbs me the most is the self censorship that was initially brought forward by TikTok and quickly spread to other platforms - all in the name of being as advertiser friendly as possible.
LinkedIn was the first platform where I really observed people losing their unique voice in favor of corporate friendly - please hire me - speak. Now this seems to be basically any platform. The only platform that seems to be somewhat protected from it is Reddit, where many mods seem to dislike LLMs as much as everybody else. But more likely, its just less noticeable
> the self censorship that was initially brought forward by TikTok
I think that’s even too soon! YouTube has had rules around being advertising friendly for longer than TikTok has existed. And the FCC has fined swearing on public broadcasts for like 50+ years.
But I do agree, we’re attributing too much to LLMs. We don’t see personal, human-oriented content online because social media is just not about community.
1. Young people (correctly) realized they could make lots of money being influencers on social media. TikTok does make that easier than ever. I have close friends who make low 6 figures streaming on TikTok (so obviously they quit the low wage jobs they were doing before).
2. People have been slowly waking up to the fact that social media has always been pretty fake. I quit 6 years ago, and most of my friends have slowly reduced how much they use it. All of the platforms are legally incentivized to only care about profit and engagement. Capitalism doesn’t allow a company to care about community and personal voice, if algorithmic feeds of influencers will make them more money.
There’s still good content out there if you know where to look. But digital human connection happens in group chats, DMs, and FaceTime, not on public social media.
I've been thinking about this as well, especially in the context of historical precedents in terms of civilization/globalization/industrialization.
How LLMs standardize communication is the same way there was a standardization in empires expanding (cultural), book printing (language), the industrial revolution (power loom, factories, assembly procedures, etc).
In that process interesting but not as "scale-able" (or simply not used by the people in power) culture, dialects, languages, craftsmanship, ideas were often lost - and replaced by easier to produce, but often lesser quality products - through the power of "affordable economics" - not active conflict.
We already have the English 'business concise, buzzwordheavy language' formal messaging trained into chatGPT (or for informal the casual overexcited American), which I'm afraid might take hold of global communication the same way with advanced LLM usage.
>How LLMs standardize communication is the same way there was a standardization in empires expanding (cultural), book printing (language), the industrial revolution (power loom, factories, assembly procedures, etc).
Explain to me how "book printing" of the past "standardized communication" in the same way as LLMs are criticized for homogenizing language.
I'm taking "same way" to be read as "authoritative", whether de facto or de jure. Basically by dint of people using what's provided instead of coming up with their own.
Everyone has the same few dictionary spellings (that are now programmed into our computers). Even worse (from a heterogeneity perspective), everyone also has the same few grammar books.
As examples: How often do you see American English users write "colour", or British English users write "color", much less colur or collor or somesuch?
Shakespeare famously spelled his own last name half a dozen or so different ways. My own patriline had an unusual variant spelling of the last name, that standardized to one of the more common variants in the 1800s.
https://en.wikipedia.org/wiki/History_of_English_grammars
"Bullokar's grammar was faithfully modelled on William Lily's Latin grammar, Rudimenta Grammatices (1534).[9] Lily's grammar was being used in schools in England at the time, having been "prescribed" for them in 1542 by Henry VIII.[5]"
It goes on to mention a variety of grammars that may have started out somewhat descriptive, but became more prescriptive over time.
2 replies →
Hits close to home after I've caught myself tweaking AI drafts just to make them "sound like me". That uniformity in feeds is real and it's like scrolling through a corporate newsletter disguised as personal takes.
what if we flip LLMs into voice trainers? Like, use them to brainstorm raw ideas and rewrite everything by hand to sharpen that personal blade. atrophy risk still huge?
Nudge to post more of my own mess this week...
It's still an editor I can turn to in a pinch when my favorite humans aren't around. It makes better analogies sometimes. I like going back and forth with it, and if it doesn't sound like me, I rewrite it.
Don't look at social media. Blogging is kinda re-surging. I just found out Dave Barry has a substack. https://davebarry.substack.com/ That made me happy :) (Side note, did he play "Squirrel with a Gun??!!!")
The death of voice is greatly exaggerated. Most LLM voice is cringe. But it's ok to use an LLM, have taste, and get a better version of your voice out. It's totally doable.
It's ironic that https://substack.com/@davebarry uses a lot of AI-generated imagery. Maybe the death of vision is not exaggerated.
I don't judge, I'm not an artist so if I wanted to express myself in image I'd need AI help but I can see how people would do the same with words.
Where are these places where everything is written by a LLM? I guess just don’t go there. Most of the comments on HN still seem human.
i think the frontpage of hn has had at least one llm-generated blog post or large github readme on it almost every day for several months now
Tbh I prefer to read/skim the comments first and only occasionally read the original articles if comments make me curious enough. For now I never ended checking something that would seem AI generated.
It’s pretty much all you see nowadays on LinkedIn. Instagram is infected by AI videos that Sora generates while X has extremist views pushed up on a pedestal.
The HN moderation system seems to hold, at least mostly. But I have seen high-ranking HN submissions with all the subtler signs of LLM authorship that have managed to get lots of engagement. Granted, it's mostly people pointing out the subtle technical flaws or criticizing the meandering writing style, but that works to get the clicks and attention.
Frankly, it only takes someone a few times to "fall" for an LLM article -- that is, to spend time engaging with an author in good faith and try to help improve their understanding, only to then find out that they shat out a piece of engagement bait for a technology they can barely spell -- to sour the whole experience of using a site. If it's bad on HN, I can only imagine how much worse things must be on Facebook. LLMs might just simply kill social media of any kind.
Ironically this post is written in a pretty bland, 'blogging 101' style that isn't enjoyable to read and serves just to preach a simple, consensus idea to the choir.
These kinds of posts regularly hit the top 10 on HN, and every time I see one I wonder: "Ok, will this one be just another staid reiteration of an obvious point?"
True, but one of the least-explored problems with AI is that because it can regurgitate basic writing, basic art, basic music with ease, there is this question:
Why do it at all if I won't do better than the AI?
The worst risk with AI is not that it replaces working artists, but that it dulls human creativity by killing the urge to start.
I am not sure who said it first, but every photographer has ten thousand bad photos in them and it's easier if they take them at the beginning. For photographers, the "bad" is not the technical inadequacy of those photos; you can get past that in the first one hundred. The "bad" is the generic, uninteresting, uninspiring, underexplored, duplicative nature of them. But you have to work through that to understand what "good" is. You can't easily skip these ten thousand photos, even if your analysis and critique skills are strong.
There's a lot to be lost if people either don't even start or get discouraged.
But for writing, most of the early stuff is going to read much like this sort of blog post (simply because most bloggers are stuck in the blogging equivalent of the ten thousand photos; the most popular bloggers are not those elevating writing).
"But it looks like AI" is the worst, most reflexive thing about this, because it always will, since AI is constantly stealing new things. You cannot get ahead of the tireless thief.
The damage generative AI will do to our humanity has only just started. People who carry on building these tools knowing what they are doing to our culture are beneath our contempt. Rampantly overcompensated, though, so they'll be fine.
I continually resist the urge to deploy my various personas onto hn, because I want to maintain my original hn persona. I am not convinced other people do the same. It is not that difficult to write in a way that avoids some tell tale signs.
> I guess just don’t go there.
How do you know? A lot of the stuff I see online could very much be produced by LLMs without me ever knowing. And given the economics I suspect that some of it already is.
Many instagram and facebook posts are now llm generated to farm engagement. The verbosity and breathless excitement tends to give it away.
There was recently this link talking about AI slop articles on medium
https://rmoff.net/2025/11/25/ai-smells-on-medium/
He doesn't link many examples, but at the end he gives the example of an author pumping out +8 articles in a week across a variety of topics. https://medium.com/@ArkProtocol1
I don't spend time on medium so I don't personally know.
I've seen AI generated comments on HN recently, though not many. Users who post them usually only revert back to human when challenged (to reply angrily), which hilariously makes the change in style very obvious.
Of course, there might be hundreds of AI comments that pass my scrutiny because they are convincing enough.
LinkedIn
I see them regularly on several subreddits, I frequent.
There are already many AI-generated submissions on HN every day. Comments maybe less so, but I've already seen some, and the amount is only going to increase with time.
Every time I see AI videos in my YouTube recommendations I say “don’t recommend this channel” but the algorithm doesn’t seem to get the hint. Why don’t they make a preference option of “don’t show me AI content”
2 replies →
Not sure if it's an endemic problem, just yet, but I expect it to be, soon.
For myself, I have been writing, all my life. I tend to write longform posts, from time to time[0], and enjoy it.
That said, I have found LLMs (ChatGPT works best for me) to be excellent editors. They can help correct minor mistakes, as long as I ignore a lot of their advice.
[0] https://littlegreenviper.com/miscellany/
I just want to chime in and say I enjoy reading your takes across HN, it's also inspiring how informative and insightful they are. Glazing over, please never stop writing.
Thanks so much!
It’s ok. Most of our opinions suck and are unoriginal anyway.
The few ones who have something important to say they will, and we will listen regardless of the medium.
Humans are evolved to spend fewer calories and avoid cognitively demanding tasks.
People will spend time on things that serve utility AND are calorifically cheap. Doomscrolling is a more popular past time than say - completing Coursera courses.
The liberal idea that the best ideas win out in the marketplace turned out to be laughably wrong.
I'd argue that they do win out, it's just not the ideas that we thought were best.
1 reply →
The marketplace is a terrible mechanism for truth-finding except for all the others. What's your proposed alternative that doesn't just relocate the problem to whoever gets to be the arbiter?
Wouldn’t say it’s a liberal idea. It was a foundational argument in jurisprudence, from Holme’s dissent in the Abram’s case.
Let's clarify, maybe the best ideas would win out in the "level marketplace", where the consumer actually is well informed on the products, the product's true costs have to be priced, and there was no ad-agencies.
Instead, we have misinformation (PR), lobbying, bad regulation made by big companies to trench their products, and corruption.
So, maybe, like communism, in a perfect environment, the market would produces what's best for the consumers/population, but as always, there are minority power-seeking subgroups that will have no moral barriers to manipulate the environment to push their product/company.
They get drowned by bots and missinformation and rage bait and 'easyness'.
Economy is shit? Lets throw out the immigrants because they are the problem and lets use the most basic idea of taxing everything to death.
No one wants to hear hart truths and no one wants to accept that even as adults, they might just not be smart. Just beause you became an adult, your education shuld still matter (and i do not mean having one degree = expert).
The problem with the “your voice is unique and an asset” argument is what we’ve promoted for so long in the software industry.
Worse is better.
A unique, even significantly superior, voice will find it hard to compete against the pure volume of terrible non unique LLM generated voices.
Worse is better.
I think there are likely a lot of people for whom their voice is not unique or an asset. Those people need to hire someone who is better at writing (assuming they're managing a business's social media). Or those people can save money and use an LLM. Whether that's a good business decision or not, I don't think human vs LLM is an apples to apples comparison, because you'd have to compare LLMs to whoever is willing to write marketing copy pro bono.
If you give an LLM enough context, it writes in your voice. But it requires using an intelligent model, and very thoughtful context development. Most people don't do this because it requires effort, and one could argue maybe even more effort than just writing the damn thing yourself. It's like trying to teach a human, or anyone, how to talk like you: very hard because it requires at worst your entire life story.
Why the f- would I train software to do my thinking and reasoning for me!?
It is not what training is, but with edgy attitude like yours, no one will want to give you their arguments.
* it writes in an imitation of your voice.
Why does this even matter? If it can say something more eloquently, in less stilted way something what I wanted to say, adding some interesting nuance on the way, while still sounding close to me - why not? I meanwhile, can learn one-two rhetorical tricks from LLMs reading the result.
1 reply →
Something that freaked me out a little bit is that I've now written enough online (i.e.: HN comments) that the top models know my voice already and can imitate it on request without having to be fed any additional context.
There's a data centre somewhere in the US running additions and multiplications through a block of numbers that has captured my voice.
In one of the WhatsApp communities I belong to, I noticed that some people use ChatGPT to express their thoughts (probably asking it to make their messages more eloquent or polite or whatever).
Others respond in the same style. As a result, it ends up with long, multi-paragraph messages full of em dashes.
Basically, they are using AI as a proxy to communicate with each other, trying to sound more intelligent to the rest of the group.
A friend of mine does this as English as second language and his tone was always misconstrued. I'd bug him about his slop, but he'll take that over getting his tone misconstrued. I get it
LOL ... in whatssap! ... sorry, we're fucked ...
<< Write in your voice.
I don't disagree, but LLMs happened to help with standardizing some interesting concepts that were previously more spread out as concepts ( drift, scaffolding, and so on ). It helps that chatgpt has access to such a wide audience to allow that level of language penetration. I am not saying don't have voice. I am saying: take what works.
> I don't disagree, but LLMs happened to help with standardizing some interesting concepts that were previously more spread out as concepts ( drift, scaffolding, and so on ).
What do you mean? The concepts of "drift" and "scaffolding" were uncommon before LLMs?
Not trying to challenge you. Honestly trying to understand what you mean. I don't think I have heard this ever before. I'd expect concepts like "drift" and "scaffolding" to be already very popular before LLMs existed. And how did you pick those two concepts of aaallll... the concepts in this world?
Apologies, upon re-reading it does seem I did not phrase those as clearly as I originally intended. You are right in the sense that the concepts existed beforehand and the words were there to capture it. What did not exist, however, was a sudden resurgence of those words due to them appearing in llms more often than note. This is what I mean by a level of language penetration ( people using words and concepts, because llms largely introduced them to those concepts --- kinda like.. genetics or pop psych, before situational comedy, projection was not a well known concept ).
Does it make more sense?
1 reply →
In a lot of ways, I'm thankful that LLMs are letting us hear the thoughts of people who usually wouldn't share them.
There are skilled writers. Very skilled, unique writers. And I'm both exceedingly impressed by them as well as keenly aware that they are a rare breed.
But there's so many people with interesting ideas locked in their heads that aren't skilled writers. I have a deep suspicion that many great ideas have gone unshared because the thinker couldn't quite figure out how to express it.
In that way, perhaps we now have a monotexture of writing, but also perhaps more interesting ideas being shared.
Of course, I love a good, unique voice. It's a pleasure to parse patio11's straussian technocratic musings. Or pg's as-simple-as-possible form.
And I hope we don't lose those. But somehow I suspect we may see more of them as creative thinkers find new ways to express themselves. I hope!
> In a lot of ways, I'm thankful that LLMs are letting us hear the thoughts of people who usually wouldn't share them.
I could agree with you in theory, but do you see the technology used that way? Because I definitely don't. The thought process behind the vast majority of LLM-generated content is "how do I get more clicks with less effort", not "here's a unique, personal perspective of mine, let's use a chatbot to express it more eloquently".
We might get twice as many original ideas but hundred times as much filler. Neither of those aspects erases the other. Both the absolute number of ideas and the ratio matter.
> In that way, perhaps we now have a monotexture of writing, but also perhaps more interesting ideas being shared.
They aren't your ideas if its coming out of an LLM
Are they your ideas if they go through a heavy-handed editor? If you've had lots of conversations with others to refine them?
I dunno. There's ways to use LLMs that produces writing that is substantially not-your-ideas. But there's also definitely ways to use it to express things that the model would not have otherwise outputted without your unique input.
counterargument: they still are your ideas even if they went through LLM.
3 replies →
I seriously doubt people didn't write blog posts or articles before LLMs because they didn't know how to write.
It's not some magic roadblock. They just didn't want to spend the effort to get better at writing; you get better at writing by writing (like good old Steve says in "On Writing"). It's how we all learnt.
I'm also not sure everyone should be writing articles and blog posts just because. More is not better. Maybe if you feel unmotivated about making the effort, just don't do it?
Almost everyone will cut novice writers and non-native $LANGUAGE speakers some slack. Making mistakes is not a sin.
Finally, my own bias: if you cannot be bothered to write something, I cannot be bothered to read it. This applies to AI slop 100%.
I hate when people hijack progressive language - like in your case the language of accessibility - for cheap marketing and hype.
Writing is one of the most accessible forms of expression. We were living in a world where even publishing was as easy as imaginable - sure, not actually selling/profiting, but here’s a secret, even most bestselling authors have either at least one other job, or intense support from their close social circle.
What you do to write good is you start by writing bad. And you do it for ages. LLMs not only don’t help here, they ruin it. And they don’t help people write because they’re still not writing. It just derails people who might, otherwise, maybe start actually writing.
Framing your expensive toy that ruins everything as an accessibility device is absurd.
I'm anon, but also the farthest thing from a progressive, so I find this post amusing.
I don't disagree with a lot of what you're saying but I also have a different frame.
Even if we take your claim that LLMs don't make people better writers as true (which I think there's plenty to argue with), that's not the point at all.
What I'm saying is people are communicating better. For most ideas, writing is just a transport vessel for ideas. And people now have tools to communicate better than they would have been.
Most people aren't trying to become good writers. That's true before, and true now.
On the other hand, this argument probably isn't worth having. If your frame is that LLMs are expensive toys that ruin everything -- well, that's quite an aggressive posture to start with and is both unlikely to bear a useful conversation or a particularly delightful future for you.
2 replies →
It's probably true that it reduces the barrier to entry, you don't refute that point in your post. You just call it cheap marketing and hype.
3 replies →
It basically boils down to "I want the external validation of being seen as a good writer, without any of the internal growth and struggle needed to get there."
3 replies →
Also that these models are being used to promote fake news and create controversy ou interact with real humans with unknown purposes
Talking to some friends and they feel the same. Depending where you are participating a discussion you just might not feel it is worth it because it might just be a bot
Sometime within the next few years I imagine there will be a term along the lines of "re-humanise," where folks detox from AI use to get back in touch with humanity. At the rate we're going, humanity has become a luxury and will soon demand a premium.
First we need to think about why we consume content? I am happy to read llm created stuff when I need to know sth and it delivers 100%. Other reasons like "get perspectives of real humans", or "resonate" ... not so much
This will get better over time, there will be ways to make LLM output more uniquely yours. I personally enjoy writing all my blog posts with LLMs, because its the only way I can bring to turn countless of notes and drafts I have while running experiments, into some public facing documentation and blog posts that I would want people to read. There are at least 5-6 years of lost ideas, thoughts and notes that I was not able to communicate the same way I can do now with LLMs. So I definitely found my voice here.
The global alignment also happens through media like tv shows and movies, the internet overall.
I agree I think we should try to do both.
In germany for example, we have very few typical german brands. Our brands became very global. If you go Japan for example, you will find the same product like ramen or cookies or cakes a lot but all of them are slighly different from different small producers.
If you go to an autobahn motorway/highway rest area you will find local products in japan. If you do the same in germany, you find just the generic american shit, Mars, Modneles, PepsiCo, Unilever...
Even our german coke like Fritz cola is a niche / hipster thing even today.
It is not a zero sum game.
I have always had a very idiosyncratic way of expressing myself, one that many people do not understand. Just as having a smartphone has changed my relationship to appointments - turning me into a prompt and reliable "cyborg" - LLMs have made it possible for me to communicate with a broader cross section of people.
I write what I have to say, I ask LLMs for editing and suggestions for improvement, and then I send that. So here is the challenge for you: did I follow that process this time?
I promise to tell the truth.
I think there's a difference between using an LLM as an editor and asking the LLM to write something for you. The output in the former I find to still have a far clearer tonal fingerprint than the latter.
And whose to say your idiosyncratic expressions wouldn't find an audience as it changes over time? Just you saying that makes me curious to read something you wrote.
I have been writing for professionally since I published my book in 2017 (with human editors): https://a.co/d/6ZFFb2z
If you want to read my unedited writing most of the articles on Hacker Noon prior to this year are "raw" https://hackernoon.com/u/azw
Most (all except the last two) of these are "raw": https://emusings.substack.com/
I never garnered a large readership. When I was the CTO at a large NGO, they literally called me "big yoda" because they considered my pronouncements so inscrutable. <shrug>
Not the GP, but I'm a millennial who leans on cultural references and has a bit of verbal flourish that I think comes from a diet of ironic, quirky, dialogue-heavy media in the early 2000s, stuff like Firefly, Veronica Mars, and Pushing Daisies, not to mention 90s Tarantino, John Cusack films, and so on.
I've never given it too much thought, it's just... the way I communicate, and most people in my life don't give much thought to it either. But I recently switched jobs, and a few people there remarked on it, and I've also recently been corresponding with someone overseas who is an intermediate-level English speaker and says I sometimes hurt their brain.
Not making a value judgment either way on whether it's "sophisticated" or whatever, but it is I think part of my personality, and if I used LLM editing/translation I would want it to be only in the short term, and certainly not as something
1 reply →
Transformation seems reasonable for that purpose. And if we were friends, I'd rather read your idiosyncratic raw output.
At some point, generation breaks a social contract that I'm using my energy and attention consuming something that another human spent their energy and attention creating.
In that case I'd rather read the prompt the human brain wrote, or if I have to consume it, have an LLM consolidate it for me.
True, I do not do that for friends
I should probably do that too. I once wrote an email that to me was just filled with impersonal information. The receiver was somebody I did not personally know. I later learned I made that person cry. Which I obviously did not intend. I did not swear or call anyone names. I basically described what I believe they did, what is wrong about that and what they should do instead.
I hear you. LLMs have been a major improvement in my life in that regard.
If someone cries about an email you sent, the problem isn’t with you.
I would be interested to see an example of a before and after on this. I do think LLMs as editors and rewriters can be useful sometimes, but I usually only ever see them used as a means to puff out an idea into longer prose which is really mostly counterproductive.
I think it can be useful as a tone-check sometimes, like show me how a frustrated or adversarial reader is going to interpret this thing I'm about to send/post.
LLMs have now robbed you of the opportunity to make your communication clearer
hahahahahaha
Here's my guess- your post reflects your honest opinion on the matter, with some LLM help. It elaborated on your smartphone analogy, and may have tightened up the text overall.
here is the entire exchange. you decide:
Me: I want to reply to this HN article:
We're Losing Our Voice to LLMs Social media has become a reminder of something precious we are losing in the age of LLMs: unique voices. Over time, it has become obvious just how many posts are being generated by an LLM. The tell is the voice. Every post sounds like it was posted by the same social media manager. If you rely on an LLM to write all your posts, you are making a mistake. Your voice is an asset. Not just what you want to say, but how you say it. Your voice is unique. It is formed from your lifetime of lived experiences. No one's voice will be exactly like yours. Your voice becomes recognizable. Over many posts it becomes something people subconsciously connect with, recognize, trust, and look forward to. Your voice provides the framework for the impression you leave in a job interview, while networking at a meet-up, or with a co-worker. Years ago I got a job thanks to my blog posts. A manager wanted my voice influencing their organization. Your voice is an asset. Your voice matures and becomes even more unique with time and practice. LLMs can rob you of that voice, and the rest of us lose something precious in the process. Having an LLM write "in your voice" is not the same. Your voice is not static. It changes with the tides of your life and state of mind. Your most impactful message may come because it was the right moment and you were in the right frame of mind. Let your voice grow with use. Let it be unique. Do not let one of your greatest assets fade into atrophy, wilted by cognitive laziness. Write in your own voice. I do not care what the linguistic remix machine juggles into being. I care what you have to say. Join my newsletter. Get my posts, best coupons, and most interesting links direct to your inbox. Email address Join © 2025 Tony Alicea. All rights reserved. tonyalicea.dev
Here is my proposed reply:
It is not a zero sum game.
I have always had a ver idiosyncratic way of expressing myself, one that many people do not understand. Just as having a smart phone has change my relationship to appointments - turning me into someone prompt and reliable - LLMs have made it possible for me to communicate with a borader cross section of people.
I write what I have to say, I ask LLMs for editing and suggestions for improvement, and then I send that. SO here is the challenge for you: did I follow that process this time?
I promise to tell the truth.
Claude: Identified typos and validated strong argumentative content.
Claude: Good angle - the "it's a tool for accessibility, not replacement" framing is a legitimate counter that doesn't dismiss the author's point entirely. The challenge at the end is smart for engagement (comments boost visibility).
Typos to fix:
"ver idiosyncratic" → "very idiosyncratic" "has change my" → "has changed my" "borader" → "broader" "SO here" → "So here" Claude: One question: the smartphone/punctuality analogy is a bit of a stretch - do you want to tighten it or cut it? It's not essential to your point.
Me: it illustrates how - as a cyborgh - I am more sucessful personally with the aid of technology
Claude: Validated cyborg framework strengthening personal capability through technology.
Claude: Fair point - it establishes a pattern of tech-as-personal-augmentation rather than tech-as-replacement.
Claude: Do you want to make the "cyborg" framing explicit in the post? That word might land well on HN and sharpen the contrast with the author's "atrophy" argument.
Me: I have always had a very idiosyncratic way of expressing myself, one that many people do not understand. Just as having a smart phone has changed my relationship to appointments - turning me into a prompt and reliable "cyborg" - LLMs have made it possible for me to communicate with a broader cross section of people.
Claude: Validated cohesive metaphor and polished prose for clarity.
Claude: That works. "Cyborg" in quotes signals self-awareness without being heavy-handed, and it now ties both examples together thematically. Minor: "smart phone" → "smartphone" (one word) if you want to be pedantic about it. Otherwise the post is clean and ready.
I don't see signs of LLM writing in your comment so I'll have to guess no.
It did some light spell checking and suggest I add one word - see above
If you didn't intentionally try and trick us, then yes, you used an LLM.
see above
Please share what you told the LLM! I can't be the only curious one.
See above
You didn't, but you've learned.
I did, but it did not have much to add
In my view LLMs are simply a different method of communication. Instead of relying on "your voice" to engage the reader and persuade them of your point of view, writing with LLMs for analysis and exploration through LLMs, is about creating an idea space that a reader can interact with and explore from their own perspective, and develop their own understanding of, which is much more powerful.
Soon, we'll be nostalgic for social media. The irony.
FWIW this prompt works for very good for me:
Your mileage may vary.
There are deterministic solutions for grammar and spellcheck. I wouldn't rely on LLMs for this. Not only is it wasteful, we're turning to LLMs for every single problem which is quite sad.
We've been losing our voice to fake, international maga influencers/Cambridge Analytica for far longer than (available) AI has existed.
The posts sounds beige and AI-generated ironically.
In any case, as someone who experimented with AI for creative writing, LLM _do not destroy_ your voice; it does flatten your voice, but with minimal effort you can make it sound the way you find reflects you thought best.
The potentially bitter pill to swallow here is that we all need to get better at critical thinking.
There's a lot of talk over whether LLMs make discourse 'better' or 'worse', with very little attention given to the crisis we were having with online discourse before they came around. Edelman was astroturfing long before GPT. Fox 'news' and the spectrum of BS between them and the NYT (arranged by how sophisticated they considered their respective pools of rubes to be) have always, always been propaganda machines and PR firms at heart wearing the skin of journalism like buffalo bill.
We have needed to learn to think critically for a very long time.
Consider this; if you are capable of reading between the lines, and dealing with what you read or hear on the merits of the thoughts contained therein, then how are you vulnerable to slop? If it was written by an AI (or a reporter, or some rando on the internet) but contains ideas that you can turn over and understand critically for yourself, is it still slop? If it's dumb and it works, it's not dumb.
I'm not even remotely suggesting that AI will usher in a flood of good ideas. No, it's going to be used to pump propaganda and disseminate bullshit at massive scale (and perhaps occasionally help develop good ideas).
We need to inoculate ourselves against bullshit, as a society and a culture. Be a skeptic. Ironnman arguments against your beliefs. Be ready to bench test ideas when you hear them and make it difficult for nonsense to flourish. It is (and has been) high time to get loud about critical thinking.
I wholeheartedly agree, I wrote about this at https://ruudvanasseldonk.com/2025/llm-interactions.
Your post reminded me how I could tell my online friend was pissed just because she typed "okay." or "K." instead of "okay". We could sense our emotional state from texting. One of those friendships you form over text through the internet. I wouldn't recommend forming these too deeply since some in person nuance is lost, we could never transition to real life friends despite living close by. But we could tell what mood we were in just from typing. It was wild.
You're absolutely right.
Here's why:
A devils advocate in me would say that this post was authored by one of LLM models creator realizing they really need more fresh meat to train on.
I'm the OP. I can attest that I am not an LLM model creator! :)
I consider myself an LLM pragmatist. I use them where they are useful, and I educate people on them and try to push back on all the hype marketing disguised as futurism from LLM creators.
There has been an explosion in verbose status update emails at my job recently which have all clearly been written by ChatGPT. It’s the fucking emojis though that drive me wild. It’s so hard to read the actual content when there’s an emoji for every single sentence.
And now when I see these emoji fests I instantly lose interest and trust in the content of the email. I have to spend time sifting through the fluff to find what’s actually important.
LLMs are creating an assymetric imbalance in effort to write vs effort to read. What is taking my coworkers probably a couple minutes to draft requires me 2-3x as long to decipher. That imbalance used to be the opposite.
I’ve raised the issue before at work and one response I got was to “use AI to summarize the email.” Are we really spending all this money and energy on the worlds worst compression algorithm?
it's not the voice. it's the lack of need to talk tough about the hard problems. if you accept what is and just babble, anything you write will sound like babbling.
there's enough potential and wiggle room but people align, even when they don't, just to align.
when Rome was flourishing, only a few saw what was lingering in the cracks but when in flourishing Rome ...
The LLM v human debate here reminds me of the now dormant "Are you living in a simulation?" discussions of previous decades.
> Social media has become a reminder of something precious we are losing in the age of LLMs: unique voices.
Social media already lost that nearly two decades ago - it died as content marketing rose to life.
Don't blame on LLMs what we've long lost due to cancer that is advertising[0].
And don't confuse GenAI as a technology with what the cancer of advertising coopts it to. The root of the problem isn't in the generative models, it's in what they're used for - and the problem uses aren't anything new. We've been drowning in slop for decades, it's just that GenAI is now cheaper than cheap labor in content farms.
--
[0] - https://jacek.zlydach.pl/blog/2019-07-31-ads-as-cancer.html
> The root of the problem isn't in the generative models, it's in what they're used for
That's like giving weapons to everybody in the world for free, and asking to be blamed for the increased deaths and violence.
No, that's like pretending the weapons weren't already available. Everyone had assault rifles for two decades, giving access to smart rifles isn't really changing anything about the nature of the problem.
I call it the enshittification fix-point. Not only are we losing our voice, we'll soon enough start thinking and talking like LLMs. After a generation of kids grows up reading and talking to LLMs, that will be only way they'll know how to communicate. You'll talk to a person and you couldn't tell the difference between them and LLMs, not because LLMs became amazing, but because our writing and thinking style become more LLM-like.
- "Hey, Jimmy, the cookie jar is empty. Did you eat the cookies?"
- "You're absolutely right, father — the jar seems to be empty. Here is bullet point list why consuming the cookies was the right thing to do..."
That's a great point — not only would I never do that but no one else should either.
Ironically I find it hard to tell whether this writing is LLM or merely a bit hollow and vapid.
I don't find that current generation LLMs output such short sentences that would start with the same prefix such as "Your voice".
You are downvoted but I actually agree with you. This blog post could have been a LinkedIn post from any "influencer", considering how generic it is.
If this is what passes for deep and insightful prose for dozens of HN drones, the situation is worse than I thought
1) People who use LLMs for generating
2) People who use LLMs for understanding
I think I'll stick to 2) for many reasons.
I'm just using the internet less and less recreationally. Except for pirating movies.
language must expand to incorporate meaning which has not yet been digested by LLM ex: 6-7
It's more that people who historically didn't have a voice now have one. It's often stupid but sometimes also interesting and innovative. Saw a channel where a university professor "I" comes to the realization she's been left-leaning/biased for decades, that her recent male students no longer dare engage in debate because of shaming/gaslighting etc. Then I click channel description and turns out it's "100% original writing". Now if it hadn't said that it would be strawman propaganda. But now it does... Not sure how to put a finger on it, there's some nervous excitement when reading these days, not knowing who the sender is, getting these 'reveal' moments when finding out whole thing was made up by some highschool kid with AI or insane person.
I am going to name my first egg baby Llmon
Well - voice is ultimately coupled to a person. LLMs thus fake and pretend being a person. There are, however had, use cases for LLMs too. I saw it used for the creation of video games; also content generated by hobbyists. So, while I think AI should actually die, for hobbyists generating mods for old games, AI voice overs may not be that bad. Just as AI generating images for free to play browser games may not be solely bad either.
Of course there are also horrible use of AI, liars, scummy cheaters and fake videos on youtube, owned by a greedy mega-corporation that sold its soul to AI. So the bad use cases may be higher than the good use cases, but there are good use cases, and the "losing our voice to LLMs" isn't a whole view of it, sorry.
Deus Ex-Machina is starting to take off ...
[Sometime in the near future] The world's starved for authenticity. The last original tweet crowned a God... then killed the kid chasing that same high. Trillionaires run continent-wide data centers, endlessly spinning up agents that hire cheap physical labor to scavenge the world for any spark of novelty. The major faith is an LLM cult forecasting the turning of the last stone. The rest of us choke on recycled ad slop.
You never had a voice to lose
We're losing our code too.
Skill becomes expensive mechanized commodity
old code is left to rot while people try to survive
we lose our history, we lose our dignity.
I think that this is imbalanced in favour of wannabe influencers, who want to be consistent and popular.
If you really have no metrics to hit (not even the internal craving for likes), then it doesn't make much sense to outsource writing to LLMs.
But yes, it's sad to see that your original stuff is lost in the sea of slop.
Sadly, as long as there will be money in publishing, this will keep happening.
Social media is a reminder we are losing our voice to mass media consumption way before LLMs were a thing.
Even before LLMs, if you wanted to be a big content creator on YouTube, Instagram, tiktok..., you better fall in line and produce content with the target aesthetic. Otherwise good luck.
Subsume your agency. Stop writing. Stop learning. Stop thinking for yourself. Become hylic. Just let the machine think everything for you and act as it acts. Those that own them are benevolent and there will never be consequences.
I think that's for the best. It was human-made slop, now it's automated slop. Can't wait for people to stop paying it attention so that it withers. "It" being the whole attention economy scam.
100% agree.
Don’t write anything with LLMs, ever. Unless having no credibility is your goal.
I'm in complete agreement with the idea that people should express themselves in their own words. But this collides with certain facts about U.S. adults (and students). This summary (https://www.nu.edu/blog/49-adult-literacy-statistics-and-fac...) reveals that:
* 28% of U.S. adults are at or below "level 1" literacy, essentially meaning people unable to function in an environment that requires written language skills.
* 54% of U.S. adults read below a sixth-grade level.
These statistics refer to an inability to interpret written material, much less create it. As to the latter, a much smaller percentage of U.S. adults can compose a coherent sentence.
We're moving toward a world where people will default to reliance on LLMs to generate coherent writing, including college students, who according to recent reports are sometimes encouraged to rely on LLMs to complete their assignments.
If we care to, we can distinguish LLM output from that of a typical student: An LLM won't make the embarrassing grammatical and spelling errors that pepper modern students' prose.
Yesterday I saw this headline in a major online media outlet: "LLMs now exceed the intelect [sic] of the average human." You don't say.
I'm in Canada and the landscape is OK. But we can definitely do better. Without properly educated men and women, I'm afraid that democracy degrades to either 1) elites stop caring about responsibilities, or 2) demagogues rallying against the elites in 1)
Even before LLMs, entire SEO industry was writing content optimized to the tee with templates.
We lose our voice based on how we use our voice.
We improve our use of words when we work to improve our use of words.
We improve how we understand by how we ask.
I’ve realized that if you say that pro AI commenters are actually bot accounts, theres not really much that can be done to prove otherwise.
The discomfort and annoyance that sentence generates, is interesting. Being accused of being a bot is frustrating, while interacting with bots creates a sense of futility.
Back in the day when Facebook first was launched, I remember how I felt about it - the depth of my opposition. I probably have some ancient comments on HN to that effect.
Recently, I’ve developed the same degree of dislike for GenAI and LLMs.
Process before product, unless the product promises to deliver a 1000% return on your investment. Only the disciplined artist can escape that grim formula.
Let's not forget to mention the rise of AI-generated video. You can't really trust any video as real anymore.
Some people, but not everyone, are abdicating their agency. Period.
And that too is an expression of their own agency. #Laissez-faire
police ignores me for 2 years and counting
test
It's a little odd for a capitalist society that values outputs so highly to also value process as much.
We've proved we can sort of value it, through supporting sustainability/environmental practices, or at least _pretending to_.
I just wonder, what will be the "Carbon credits" of the AI era. In my mind a dystopian scheme of AI-driven companies buying "Human credits" from companies that pay humans to do things.
For those of us not constantly online, we're doing just fine.
I suppose when your existence is in the cloud, the fall back to earth can look scary. But it's really only a few inches down. You'll be ok.
We have a channel at work where we share our experiences in using AI for software engineering.
Predictably, this has turned into a horror zone of AI written slop that all sounds the same, with section titles with “clever” checkbox icons, and giant paragraphs that I will never read.
[dead]
[dead]
[dead]
[flagged]
"Over time, it has become obvious just how many posts are being generated by an LLM. The tell is the voice. Every post sounds like it was posted by the same social media manager."
I'd love to see an actual study of people who think they're proficient at detecting this stuff. I suspect that they're far less capable of spotting these things than they convince themselves they are.
Everything is AI. LLMs. Bots. NPCs. Over the past few months I've seen demonstrably real videos posted to sites like Reddit, and the top post is someone declaring that it is obviously AI, they can't believe how stupid everyone is to fall for it, etc. It's like people default assume the worst lest they be caught out as suckers.