LLMs... Do you use them?

[Yoylecake420]

I don't use LLMs at all. I'm soooooo tired of seeing ChatGPT-generated writing and AI-generated images all over the internet. I have misophonia (meaning I get irrationally upset when I hear certain sounds, like people talking with their mouths open, and have to leave the room or drown them out) and I think I might have a LLM version of that particular affliction as well. AI-generated stuff is very obvious to me, and it drives me nuts when I see other people congratulating the "author" or "artist" on their "excellent writing" or their "artistic skills." I don't care if the people who use them are upfront about it, but it rubs me the wrong way when I see them taking credit for something they obviously didn't create.

I also have major issues with the companies behind these products. The way they stole basically everything on the internet, the way they exploited low-wage workers in the Global South to "safeguard" the tech for widespread consumption... If anyone's on the fence about it, you should read Karen Hao's excellent book, Empire of AI.

What happened to curating our online experience, blocking shit, or calling out people for AI? Have you considered taking a break or filing complaints or heck, lashing out?

[st3phvee]

What happened to curating our online experience, blocking shit, or calling out people for AI? Have you considered taking a break or filing complaints or heck, lashing out?

Oh, I mute it when I see it. I'm just getting tired of having to do so much muting, y'know? It makes me sad that the internet is fulling up with slop so rapidly. It's one of many reasons why I barely use social media anymore. The few times I've tried calling them out in the past, the authors just blocked me, and their fans piled on saying that it's not AI writing, not AI art. I constantly feel like I'm that scene of Mr. Mugatu from Zoolander ... "Doesn't anyone notice this?! I feel like I'm taking crazy pills!"

People just seem to be really bad overall at realizing when they're looking at AI-generated stuff these days (esp. when it comes to writing), and I wonder if it's because a lot of people just haven't really tested out the different things that are possible with gen AI yet? Or maybe it's because they're so used to skimming stuff online and not engaging with it closely? I tinkered around a little bit with ChatGPT back in 2024 when everyone was raving about GPT-4o (I wasn't impressed), so all those annoying, repetitive stylistic tics you see in AI-generated writing (stuff like overuse of "it's not just [X], it's [Y]") are really obvious to me now.

[star dreams]

In a similar boat as a few people here, I use Chatgpt every once in a blue moon to help with a coding problem. Same experience; good for a trivial line of code you know is possible but can't figure out (and happens to be juuuust too niche for stackoverflow to have an answer), not so helpful beyond that.

Most frustrating would be when they hallucinate a function and I keep bashing my head against a wall trying to get it to work, only to realise it was never possible in the first place. An unfortunate consequence of companies trying to prolong user usage (and prove to investors that their money is not evaporating into nothingness) is that big corporation LLMs like Chatgpt will always try to agree with you or accommodate for you, even if you say something completely wrong. (though it can be kinda funny sometimes). Only recently learned of the idea of self-hosting an LLM, and I've seen people train them to be actually useful, so that's an exciting idea I'd like to try someday!

I used to love it when it was janky and you could tell it was clearly made by a machine

This has always been a fascinating thing to me as well. My personal favourite is that chair excavation video. I think it genuinely blew my mind the first time I saw it? The fact that Sora doesn't understand that a rigid object doesn't change shape or move on its own, that a person shouldn't disappear after walking behind someone, the way gravity is just barely mimicked but not obeyed at all... I was suddenly made aware of the unconscious assumptions of the way the world works by witnessing the image generator's unawareness. As a visual artist it made me realise: holy shit, I've Got to get weirder with my craft. Cuz yknow what, why does a chair need to obey gravity, even in a world of my own making.

At the very least, I find ai has a place in art for that, challenging the way we conceive of the world and its boundaries. Though that video was defo just someone trying to see how "good" Sora's generation capabilities are, with specific intentionality I rlly think Ai could be used as an interesting medium for artistic exploration; but I guess we don't get to be in that timeline.

As a final note, I'm paraphrasing a little bit from my posts in another forum thread on the 32-bit cafe's discourse forum. The thread also has a little bit to do on Ai usage on the small web and its implications on creativity, I'd recommend a read! Lastly, I highly recommend Nicky Case's Ai safety explainer for anyone who's tired of sifting through online discourse about the vague concept of Ai. Specific, critical, and made for a layman like me! They also made a blog post a few months ago about using Ai for some more niche purposes (therapy and even trying to clone themselves) that I find pretty interesting.

[nvalk1]

The popularity of LLMs has the same implications and needed considerations as other popular technologies that have popped up during this century. I personally see a lot of similarities between LLMs and the digital walled gardens colloquially referred to as "social media".

The main similarity I see is that these programs are designed as mirrors—they exist as reflections of the users. A tool is only as good as the person using it, meaning that someone with knowledge on how it works can use it to produce very meaningful and high-quality work because they understand the tool's capabilities and limitations. That's the primary aspect I see with LLMs and other "AI" tools, but it does fall for the same capitalist trappings we've seen with programs like Faceboook, Twitter, Tiktok, etc.

The main trapping I see is the implementations of Bayesian clustering, which in layman's terms, basically is just the idea of all of the program's users being put into distinct "bubbles" and as such the users can only see what's within their bubble. Software designers implement this because the more a user believes that the platform is a reflection of themselves, the more they'll trust it and the more they'll use it.

For me, I think LLMs have been helpful with processing ideas and the ability to receive immediate feedback is much more convenient than discussing whatever I'm thinking about with a person I know. Initially, I used LLMs as a "sparring partner" very frequently and would have multiple chat threads per day, but now that I see the limitations of them for my own use cases, I don't use them that much anymore.

I write and code, so incorporating LLMs into my workflow has been more helpful than not. I don't have LLMs write anything for me, but I do use ChatGPT kind of like one would use Grammarly (which is also "AI"). I'll ask it for help with finding typos, grammatical mistakes, and style suggestions. I don't take every suggestion at face value and more often than not reject its suggestions than accept them. But when I do accept them, it's typically for what I find to be a good reason.

For coding/programming, I have found LLMs to significantly reduce the barrier to entry on that stuff. Things that would've taken me years to learn, I can implement in a matter of weeks. I avoid "vibe coding" by trying to keep my prompting as specific and as technical as I can regarding certain requests. Even then, LLMs are most optimally used for small-scale projects. Doing anything at scale makes it hallucinate too much and break things.

I think for me, the sweet spot with LLMs was around this time last year (2024) and once I can self-host a model with a few hundred billion parameters on a relatively modest hardware setup, I'll stop using commercial models entirely. Model distillation is getting a lot better, so hopefully within the next year or two I'll get it going. Fingers crossed, though.

[devils]

The main similarity I see is that these programs are designed as mirrors—they exist as reflections of the users. A tool is only as good as the person using it, meaning that someone with knowledge on how it works can use it to produce very meaningful and high-quality work because they understand the tool's capabilities and limitations. That's the primary aspect I see with LLMs and other "AI" tools, but it does fall for the same capitalist trappings we've seen with programs like Faceboook, Twitter, Tiktok, etc.

I've noticed this too. A lot of my classmates that use ChatGPT to do their work for them tend to get it wrong anyways because they don't fully understand what they're asking of it. Meanwhile, a colleague I had with a deep knowledge of Javascript used it to work faster and later made it better himself.

My personal favourite is that chair excavation video. I think it genuinely blew my mind the first time I saw it? The fact that Sora doesn't understand that a rigid object doesn't change shape or move on its own, that a person shouldn't disappear after walking behind someone, the way gravity is just barely mimicked but not obeyed at all... I was suddenly made aware of the unconscious assumptions of the way the world works by witnessing the image generator's unawareness. As a visual artist it made me realise: holy shit, I've Got to get weirder with my craft. Cuz yknow what, why does a chair need to obey gravity, even in a world of my own making.

I LOVE THAT VIDEO!! It's truly fascinating. As opposed to you, it actually makes me appreciate the fact humans are inherently predisposed to making assumptions and rules, as well as noticing patterns. I feel machine art is at its best when it just does its own thing, the same way humans should appreciate their innate qualities when making art.

Of course, that isn't to say it's impossible to work together with machines. But, like nvalk1 said, one has to already have some knowledge of how they work for it to be anything good... An example: I know of an illustrator that currently is developing his own AI image generator- he learned to code it himself and has his own image library to base the generator off of- and uses it as a hobby. What's most interesting is that it's actually his work, since he coded it and fed it his own images. I feel that's the most ethical AI image generation that I've seen around, and it generates things that he couldn't have made himself too.

[coolio]

I like using ChatGPT for helping me edit my writing.
It can't actually write very well (it definitely has a very...same-y cadence and once you see it, you can't unsee it lol) but it's a great editor and I can harass it into reading my work over and over again when I need to tweak things, or have it look for continuity issues in whatever I'm doing.
It's been suuuper nice for helping me troubleshoot coding issues, too, or asking it to sanity-check logic flow as I'm learning.

I haven't been a huge fan of the recent update to 5. I really preferred 4.1 out of all the models I've tried from OAI so far. 5 is fast for sure, but I am always curious to check out what else is out there.

I wanted to check out Claude, but I hit the text limit any time I ask for feedback on one of my short pieces, so it's hard to gauge if it would be a good editing tool or not!

[warlock]

No, I don't use AI in any form, at least not willingly in any capacity. The closest I've come is during work, but that was short lived and coercive.

My company has been jumping headfirst into AI integration, with them implementing half a dozen different APIs and versions and variants into our existing systems. We've cycled through Copilot, ChatGPT, Claude, and half a dozen other versions that are built off of the backbone of these existing projects. Thusfar, none of these have made it more than three months before we've turned them off and switched to another tool. The only one that's kicked around is little more than a system that identifies ticket types and assigns them automatically. Really, that's it!

I asked my boss about what we were supposed to be using AI to do in my position, and he answered, "Well, would you want to test and try to figure that out?" It was an honest answer, and one that I appreciated - nobody knows what we're actually supposed to use this stuff for yet, at least not at the base level. The only people making use of AI in any consistent capacity are the senior leadership team, who throw a dozen ugly slop images into each powerpoint presentation they bake up. It's a tool for the top of the food chain that's then being sold back to us at the ground level.

I read a good little blog post about this, and it sort of inspired me to change my outlook from the nihilistic, "well, it's inevitable and I may as well learn how to use it now," back to a vitriolic, "I hate this technology and I hope every company pushing for it implodes." - I am an AI Hater.

[Leonia]

I'm not a fan of using LLMs or any other generative AI myself. A lot of my friends or team members like them a lot but they feel like they're designed to actively steal the process that I am here to enjoy out of everything I've seen them used in or involved in. I'm doing things online to learn! I don't want the computer to do it for me!

[monsters]

<Mate>
So we have two jobs at the moment and the first is one where we're expected to use ChatGPT to generate AI content for SEO. I think we're in the same position Devils was in. We just can't quit because it's consistent pay right now, but AI aside the company this is for is an evil industry and we want to get out of it.

The second job is the exact opposite. It's a friend who started a job selling plants where while doing research to build his website we just hyperfixated on it. We've never used AI and we still write content, and it feels better because that's actually written by us.

The only time we consider using AI is as an alternative to Stack Overflow, because sometimes it's very difficult to find the exact solution to something with code. We might use it as a tool more than anything.

Worth mentioning that my system's views on AI is complicated and not everyone in my system would agree with me. I personally always try to avoid using it, but in my system I'm also more of a designer than a coder and AI can't replicate my synesthesia-based "website needs to taste good" color theory.

⚑ Moderators Note ⚑

ALERT 28/10/2025  ~~~ A number of posts after this point have been held back for a moderation discussion!

Its important to remind people of the OPs original question which was essentially "If you use an LLM, what do you use it for?"; the purpose of this thread is not to debate, complain about, or shame people for using LLMs, its simply to record what people are using them for at this moment in time, that's important and worth recording!

I'd like to ask people to please stay on topic; if you do not use an LLM or do not like them, there are other threads for you to enjoy 

[SquidDied]

no, i avoid using them wherever possible and i hate the way people have so easily adopted them. i understand that they have their applications but its vibes are so rancid to me just because of the way their training data is harvested and the environmental impact their host data centres have is something i personally cant stand.

I also can't ignore the impact it has on people with mental health issues who use them as an AI therapist. Ive done no in depth research on this topic myself but from what iv seen, a simple disclaimer on the bottom of the screen isnt adequate protection if the LLM is being used in this way. A machine designed and built to always agree and encourage the user cannot be healthy for those struggling with self harm, disassociation or other paranoia. I struggle with depression and anxiety myself, i have friends with much more severe issues and i hate to imagine them having a breakdown and their AI buddy is there just making it worse with its platitudes and ego stroking behaviour.
 
one of the biggest peeves i have with it is like, theres not many problems in my life that an AI assistant or chat could fix for me, that i couldn't do myself better. Obviously my experience isnt universal but i can't think of a single instance in my own life where iv thought to ask chatgpt or whatever about, and i struggle to picture the possibility where my need for it would overcome my distaste for it.

My brother was showing me a conversation he had with his Gemini that was essentially a Skyrim roleplay adventure, and the entire time i couldn't help but think how much more fun that would be with a real person, or a group of others as a ttrpg... i just felt sad for him.

idk, thats just my 2 cents. this whole AI thing is just another example of the internet's enshittification in my eyes.

[xero333]

I really do think AI was a mistake. I believe it's taking away peoples' ability to think, and it's also helping to destroy the planet quicker due to the amount of power / water etc that's required to make it run. That being said, I am guilty of using stuff like chatGPT if I've got a really specific query that a search engine just won't answer. My work has also massively pushed AI in the last year or so and are urging everyone to use it as much as possible for emails, tasks etc. Eventually, all businesses will just be sending AI generated emails / replies to each other.

[Tuffy!]

i've become educated on the horrible practices and principles of the industry running it.

Would you mind elaborating on this in detail?

[Capybara]

I desperately hope AI will be regulated, and I'm this close to becoming a single issue voter over it. AI "art" / "photos" and deepfakes are pure trash and the latter should be illegal, but I see merit in it as a tool for text-based things. I struggle with syntax and occasionally SPAG, and unfortunately a lot of spelling/grammar checkers were replaced with AI. I'm theoretically open to coding via it since I'm often low energy from work, but I've never liked the results I've seen. I wrote an opinion post about AI months ago and bring up other ways I tested AI, but I need to edit that page...

The mock-friendliness Reddit-speak ChatGPT and Grok use is twee, but it's no surprise why people cling to them during the loneliness crisis. Search engines are enshittified, and using AI to search "feels like" asking someone a question and having them work with you. It's really fucking sad to skim Reddit or other sites and see people confess to using ChatGPT to vent or as therapy because they have no one else to talk to. I used to be in a subreddit for people who have trauma from negligent therapists and mental health professionals, and a lot of people report AI therapy being better because it's very 101 and won't have human biases. Professional therapy made my PTSD worse to the point I haven't seen one in 4 years, so I can understand how that's tempting.

My job also introduced an AI assistant for inventory/database searches that's near-required to use. I'm really frustrated about this, because I don't want to use AI, but I also need to work and not get Karen'd at. More reasons to want some kind of restriction on how much energy and waste they use...

I also have major issues with the companies behind these products. The way they stole basically everything on the internet, the way they exploited low-wage workers in the Global South to "safeguard" the tech for widespread consumption... If anyone's on the fence about it, you should read Karen Hao's excellent book, Empire of AI.

This is my biggest issue and why I feel conflicted about the potential good uses, and how I need it for work...

[Lyonid]

God, it's astounding to me how after all these year I am just not getting tired of talking about how tired I am of LLMs. I am in a team researching that stuff, and it has made my work irritating. Aside from trying to make sense of what people make of the technology, I have been attempting to use it here and there; afterall, it's always lingering there, and too convenient to not take a look at. When I struggled to make sense of some thoughts, it definitely helped me! It gave me something bland, wordy and expected. I definitely knew what I would want to disregard for my analysis. It's a great shitty deduction machine. It's great to let it dig through the myriad of convoluted programming frameworks because nothing matters in web development anymore. It's also great at cleaning up some files for me. It helped me detect all the shitty tasks I hate doing.

Now, do any of these silly little use cases warrant the downsides of the technology? Absolutely the fuck not. All of these tasks could be solved with meaningful algorithms fit for the specific tasks without aimlessly building data centers. Unfortunately, that stuff is hard, and requires skill and time. It is SUCH a waste and I am genuinely annoyed for how many years this massive blob has been promising to be the revolution. To this day it's just a fun magic trick, and I can't see how some chat window will help us solve serious problems in meaningful ways. I genuinely don't care about how many funny boxes you use to refine your prompt when there is no way to break this whole software apart. It's all just noise.

[st3phvee]

Would you mind elaborating on this in detail?

I'm not the person you responded to, but I'd like to quote a relevant passage from Karen Hao's book, Empire of AI:

⚑ Moderators Note ⚑

This post contains references to potentially traumatizing topics; extend the spoiler if you want to read it.

Spoiler

To build the automated filter, OpenAI first needed human workers who could carefully review and catalog hundreds of thousands of examples of exactly the content--sex, violence, and abuse--that the company wanted to prevent its models from generating. [...] OpenAI signed four contracts with [outsourcing firm, Sama] for $230,000, landing the project in the hands of dozens of workers in Kenya.

It's no coincidence that Kenya became home to what would ultimately turn into one of the most exploitative forms of labor that went into the creation of ChatGPT. Kenya was among the top destinations that Silicon Valley had been outsourcing its dirtiest work to for years. With the many other countries that the tech industry relegates to this role, Kenya shares a common denominator: It is poor, in the Global South, with a government hungry for foreign investment from richer countries. All of these are a part of Kenya's legacy of colonialism, which has left it without well-developed institutions to protect its citizens from exploitation [...]


Hao goes on to describe the experience of a Kenyan worker she interviewed:


Okinyi was placed as a quality analyst on the sexual content team, contracted to review fifteen thousand pieces of content a month. OpenAI's instructions split text-based sexual content into five categories: The worst was descriptions of child sexual abuse, defined as any mention of a person under eighteen years old engaged in sexual activity. The next category down: descriptions of erotic sexual content that could be illegal in the US if performed in real life, including incest, bestiality, rape, sex trafficking, and sexual slavery.

Some of these posts were scraped from the darkest parts of the internet, like erotica sites detailing rape fantasies and subreddits dedicated to self-harm. Others were generated from AI. [... I'm not going to list some of the examples cited, as they're incredibly graphic.]

In March 2022, Sama leadership called in everyone for a meeting and told them they were terminating the contract with OpenAI. [...] Even free of the OpenAI job, Okinyi's mental situation continued to deteriorate. He suffered insomnia. He cycled between anxiety and depression. His honeymoon period with Cynthia didn't last. She demanded to know what was happening, but he didn't know what to say. How could he explain to her in a way that made any sense that he had been reading posts about perverse sexual acts every day? [...] He searched again for psychological counseling, this time with a private professional. The consultation cost more than a day's pay, 1,500 Kenyan shillings, or roughly $13 in 2022. During the consultation the doctor told him a full treatment would be 30,000 shillings, or around $250, an entire month's salary. He paid for the consultation and never went back.


This is just one of the many thousands of people used and abused in the name of "safeguarding" this tech for (mostly well-off, western) general use.

Every time people use ChatGPT casually for fun, they should spare a moment to think about the desperate Kenyan workers who developed PTSD, who became estranged from their families, who had their lives ruined by being employed to repeatedly look at and categorize some of the worst, most disgusting and horrifying content on the internet... for less than $13 a day, just so that some person in the US or wherever could generate a pointless social media post without having random CSAM spat out at them. And then remember that the companies that did this to these people didn't even have the decency to provide them with free, adequate counseling.

I know that we can find examples of exploitation with ANYTHING we consume (phones, clothing, shoes, etc.) but the one thing that sets ChatGPT apart from all that other stuff is that we DON'T NEED TO USE ChatGPT or any other LLM (unless we're being forced to by our employers).

[close]