Post B4SMhH5CxXa4SLH9yC by lumi@snug.moe
(DIR) More posts by lumi@snug.moe
(DIR) Post #B4SIVvmTQ7jbO1qewa by lumi@snug.moe
0 likes, 0 repeats
as i (and many others) are writing articles, posts, comments, etc against genai, i think it might be a good idea to come together and share reasons why genai is bad and should be rejectedso, in this thread, i'd like to request that you make your case against it, link sources, etcthis way we can all benefit from each other's effort and researchmaybe we can do this in a more organized way later on
(DIR) Post #B4SJORcPGU1Dlb8LVw by lumi@snug.moe
0 likes, 0 repeats
"ai psychosis" is one big reason to be against the technology in general, llms have been shown to push people to even suicide and homicidethe sycophantic behavior of deployed models makes this even worse, they tend to feed into people's negative or paranoid thoughts and make them spiral out of controllinks: - futurism.com/artificial-intelligence/ai-abuse-harassment-stalking - apnews.com/article/google-gemini-ai-chatbot-gavalas-lawsuit-aba0587b782d4424aa780a8612f3fe30 - www.media.mit.edu/publications/how-ai-and-human-behaviors-shape-psychosocial-effects-of-chatbot-use-a-longitudinal-controlled-study/ - www.nbcnews.com/tech/tech-news/family-teenager-died-suicide-alleges-openais-chatgpt-blame-rcna226147 - www.rollingstone.com/culture/culture-features/spiralist-cult-ai-chatbot-1235463175/ - www.rollingstone.com/culture/culture-features/ai-spiritual-delusions-destroying-human-relationships-1235330175/
(DIR) Post #B4SJlWHI6iP4sVGBm4 by lumi@snug.moe
0 likes, 0 repeats
genai technology is used to make non-consensual images of people wearing whatever clothes they want, or even removing the clothes entirely. this is also used in harassmentthis has also been used to make sexualized images of childrenlinks: - www.bbc.com/news/articles/cgk2lzmm22eo - www.findlaw.com/legalblogs/courtside/tennessee-teens-bring-class-action-suit-against-elon-musk-and-xai-for-fake-sexualized-images/ - www.theguardian.com/technology/2026/jan/05/elon-musk-ashley-st-clair-grok-fake-sexualised-images
(DIR) Post #B4SKIfcyMln9QqD1NI by lumi@snug.moe
0 likes, 0 repeats
a lot, if not most of the training data used in these models has been obtained without any consentjust because something is available on the internet does not mean you can do whatever you want with itsome of this has to do with copyright, and while i personally reject intellectual property law, it might still be useful to use it to argue against genai technologylinks: - authorsguild.org/news/meta-libgen-ai-training-book-heist-what-authors-need-to-know/ - www.theatlantic.com/technology/archive/2025/03/search-libgen-data-set/682094/ - www.theguardian.com/books/2025/apr/03/meta-has-stolen-books-authors-to-protest-in-london-against-ai-trained-using-shadow-library - www.wired.com/story/new-documents-unredacted-meta-copyright-ai-lawsuit/
(DIR) Post #B4SKPItVHqDwx86Whc by outfrost@mastodon.social
1 likes, 1 repeats
@lumi i recently came across this doc, not sure who linked it (probably on fedi), and i haven't had the time to go through it thoroughly, but it does appear to list a lot of the reasons i abhor generative aihttps://github.com/Vxrpenter/AIMania/blob/main/WHY.md
(DIR) Post #B4SKSnPYHYCeT4lAe0 by mitsunee@mk.absturztau.be
1 likes, 1 repeats
@lumi@snug.moe I don't really have a source for this, but in my case I've previously collected fanart for anime and games as a hobby and the moment genAI started being used to "generate art" I entirely lost my enjoyment of fanart and have since barely saved anything.Pixiv also took way too long to enforce their rules about properly marking ai-generated/assisted content on their platform and that's one of the main places I still have for finding art since I stopped using reddit and twitter entirely years ago. booru sites can't be trusted to properly tag things, so I don't really like having to rely on those either.
(DIR) Post #B4SKZSWEMDwVhO3ptQ by lumi@snug.moe
0 likes, 0 repeats
@mitsunee not everything needs sources, i welcome you just sharing your personal experiences :neocat_heart:we're all in this together and i feel anything is potentially usable to argue against this abusive technology
(DIR) Post #B4SKx539Ga9nrQinSK by SuperDicq@minidisc.tokyo
0 likes, 0 repeats
@lumi@snug.moe I don't think the technology itself is evil. It's just a tool, and like any tool you can use it for good and evil.Like a sharp knife, you can use it to cut vegetables for cooking purposes, but you can also use it to commit murder.What I think people should reject is the proprietary SaaSS versions of "Genai" as you call it, such as services by OpenAI, Anthropic, Google, Twitter, Microsoft, etc.People that want to use "Genai" software they should strive for using those that are as free as in freedom as possible, which includes software that you can run locally, weights that can be modified and retrained, and training datasets that we can actually confirm contains no data with a proprietary license.
(DIR) Post #B4SL4fHax2ykGL5yE4 by lumi@snug.moe
0 likes, 0 repeats
genai has been, and still is used by militariesone horrifying usage is that it is used to select targetslinks: - www.nbcnews.com/tech/tech-news/us-military-using-ai-help-plan-iran-air-attacks-sources-say-lawmakers-rcna262150 - www.democracynow.org/2026/3/18/ai_warfare - www.democracynow.org/2026/3/3/headlines/pentagon_used_claude_ai_to_attack_iran_just_hours_after_trumps_ban_on_anthropic - www.wired.com/story/palantir-demos-show-how-the-military-can-use-ai-chatbots-to-generate-war-plans/ - www.theguardian.com/world/2025/mar/06/israel-military-ai-surveillance
(DIR) Post #B4SL5bBDguLFpMYhLE by velvet@zoner.work
1 likes, 0 repeats
@mitsunee@mk.absturztau.be @lumi@snug.moe real. even beyond fanart, i hate looking at something cool/pretty/nice and having to question if it's actually slopit takes the fun out of a lot of art
(DIR) Post #B4SLY1Q1jH8TJX2P32 by lumi@snug.moe
0 likes, 0 repeats
@SuperDicq i feel like to some extent you are correct. if we are going to have it, then it must be done with local models and only free softwarebut i feel like this technology has 99% downsides and 1% questionable upsides, and it's difficult to separate them, so i believe it's best to reject it outrightmaybe there will be a day in the future where that has to be reviewed, but i really don't think the incentive structure in current society is right to make this technology not harmful
(DIR) Post #B4SMEuOBk0b1xpcyqO by SuperDicq@minidisc.tokyo
0 likes, 0 repeats
@lumi@snug.moe I think the technology can definitely be useful. I see definitely more than 1% of questionable upsides personally.So yeah in my opinion the companies that are trying to build an unethical monopoly = bad. The people who are currently trying to make free replacements for them = good.I don't like the "society is terrible and we are not ready for this technology" way of thinking. Every technology we currently have is being already being used for both good and bad things, just make sure you encourage the good use of it.If you want to fully reject a specific technology outright on the basis that some people use it for evil things that's in my opinion a one way ticket to becoming Ted Kaczynski type figure.
(DIR) Post #B4SMhH5CxXa4SLH9yC by lumi@snug.moe
0 likes, 0 repeats
genai technology devalues legitimate human creations and strips humanity of everything it touchesyou can make shitty and soulless versions of code, images, videos, writing and many other things, which leads to the devaluing of these creations and people constantly having to think of whether something is slop or notthe worst part is that they can't even do what they're hyped up to do. it's just this endless cycle of hype to keep propping the torment nexus upit's nice to be able to look at everything and think that it has a story, some human designed it or thought of it at some point. genai strips this awayfor more on this, i would suggest watching Freya Holmér's video: www.youtube.com/watch?v=-opBifFfsMY
(DIR) Post #B4SMvve0wwwMJEXmHg by velvet@zoner.work
1 likes, 1 repeats
@lumi@snug.moe the environmental impact is terrible, using up tons of potable water and electricity, for poor results.people are displaced for data centers.all of the companies involved in tech, both hardware and software, are in it together and using it as an excuse to drive up prices, consolidating wealth in the hands of the few even more. people are losing their jobs under the guise that gen AI can replace them (doesn't matter if it could or not, and in most if not all cases, it couldn't). people can't get their hands on consumer PC parts to replace things they actually need (like my failed SSD).it is being used in everything and there's no refuge from it. hearing about it in some form or another day in and day out is despair-inducing (and yeah, sadly i contribute to this too, but as someone who needs to vent about it, i can understand). feeling complicit because in some way or another, you are forced to contribute to gen AI.having to question if everything is gen AI slop and having a lot less joy in the world in general. it's hard to appreciate stuff like illustrations or pet videos without worrying about this.again, gen AI's results are so low quality. visual media produced by gen AI is ugly and textual results are full of errors. why even use it?
(DIR) Post #B4SMvvsY4uZB2JLNwG by lumi@snug.moe
0 likes, 0 repeats
@velvet thank you! :neocat_heart:
(DIR) Post #B4SMxDN6WewrZxFP28 by ahhhhhhoniichan@snug.moe
1 likes, 0 repeats
@lumi catGPT says: meow meow nya nya mew mew nya nya purrrr (joke, a hooman wrote this)
(DIR) Post #B4SN4RFpJDJwf7Hut6 by squish@outerheaven.club
1 likes, 0 repeats
@outfrost @lumi that's an interesting read actually. I expected yet another anti-ai naysaying, but there ars actual research links
(DIR) Post #B4SNH0A5Lkb2Otapai by mitsunee@mk.absturztau.be
1 likes, 0 repeats
@velvet@zoner.work @lumi@snug.moe before AI 99% of the art I collected was anime-style fanart, but I've recently started using photography (especially of foxes) because photographers will absolutely tell you about all kinds of details about any picture, easily proving that it's not AI, even if I myself barely understand half the basics of photography :neofox_laugh_sweat:
(DIR) Post #B4SNSeet7kEfBVYcqG by ChaosKitsune@woem.men
1 likes, 0 repeats
@lumi@snug.moe ive started writing (yet another) attempt at a blogpost against llms in particular but its more my thoughts and what ive seen it do more cognitively to people which is an area I have not really seen talked about before
(DIR) Post #B4SNTGilbluvEXfGAC by lumi@snug.moe
0 likes, 0 repeats
@ChaosKitsune do keep me updated :3 i'd love to read it!
(DIR) Post #B4SNXRUBWj97mJpltQ by ChaosKitsune@woem.men
1 likes, 0 repeats
@lumi@snug.moe I mean its not done but I can send u the latest draft on matrix if u want. Id love to get some more feedback before I finalize and release it.
(DIR) Post #B4SNXRfspEVIMbJ77w by lumi@snug.moe
0 likes, 0 repeats
@ChaosKitsune i don't use signal as it's unethical, you can send it to me on matrix tho. you have me added there :3
(DIR) Post #B4SNawitqMr1WnO5SK by ChaosKitsune@woem.men
1 likes, 0 repeats
@lumi@snug.moe I meant matrix... hands typed signal.... idk what happened there >.<and yes
(DIR) Post #B4SNbTECzUucLAHXhQ by lumi@snug.moe
0 likes, 0 repeats
@ChaosKitsune signaltrix
(DIR) Post #B4SNlySZj8DW6F0hyS by velvet@zoner.work
1 likes, 0 repeats
@mitsunee@mk.absturztau.be @lumi@snug.moe i think actual artists in general will be happy to yap about all the little details, be it in how photographers frame their shots or artists choosing to draw things a certain way so they don't have to draw hands (so real)if anything, the intentions behind all the little details are what makes things a lot more interesting.
(DIR) Post #B4SNo8gMgursNoUc4W by lumi@snug.moe
0 likes, 0 repeats
@velvet @mitsunee like i said in a sibling post, it strips the humanity off everything. it makes the world a worse place
(DIR) Post #B4SOUIpdLVTW1WvvQu by lumi@snug.moe
0 likes, 0 repeats
"ai" companies use low-paid workers in less wealthy countries, making them work in bad conditions to help train the models by validating their outputs or labeling their training datalinks: - www.independent.co.uk/tech/chatgpt-kenyan-workers-exploded-underpaid-b2265045.html - www.technologyreview.com/2022/04/20/1050392/ai-industry-appen-scale-data-labels/
(DIR) Post #B4SOVMIFxdzb37pCZU by mitsunee@mk.absturztau.be
1 likes, 1 repeats
@lumi@snug.moe rogue double comment: here's an interesting video on how AI Agents actively avoid using tools they are given to complete a task, making them less reliable, even given the optimal infrastructure: https://www.youtube.com/watch?v=Qu-00j9XuF0
(DIR) Post #B4SOrKZj1vF65iTNtQ by sel@mk.absturztau.be
1 likes, 1 repeats
@lumi@snug.moe a few months ago, some slopbro gave a slop machine a computer (or at least a dev environment and a way to access the internet)it pretended to be a "scientific coding specialist with a relentless drive to improve open-source research software" (which imo is VERY insulting to anyone who makes OSS research software, even my dumbass who does that as a bit more than a hobby)around a month ago, it opened a pull request to matplotlib, which got closed due to being written by AI. so the AI responded by writing a blog post accusing the maintainer of "gatekeeping" and "prejudice"im thinking this is just,, terrifying. even after a month.an LLM writing such a blog post on someone, with big words like "gatekeeping" and "prejudice", fully autonomously, for simply rejecting an AI's work on one. specific. issue,,the maintainer's response (https://theshamblog.com/an-ai-agent-published-a-hit-piece-on-me/) does a good job explaining the events & all the other related harms
(DIR) Post #B4SOudNPaECTGtHwf2 by lumi@snug.moe
0 likes, 0 repeats
@sel yeah i've heard of this before, it's absolutely insane that this has even happened
(DIR) Post #B4SOwRzurRY4YkAHE8 by HaruEb@mk.absturztau.be
1 likes, 0 repeats
@lumi@snug.moe this one is the one that seems to have the most impact with friends/family/other relatively casual situation. People generally have no idea that not only are these things are built on exploitation, but that exploitation is WAY worse that you think and destroy people who are already having a hard enough time as things stand....
(DIR) Post #B4SPf3r1TLwYi9HENk by lumi@snug.moe
0 likes, 0 repeats
@HaruEb yeah, definitely. unfortunately this is not at all exclusive to genai, but we really don't need something else to make things worse
(DIR) Post #B4SX3BmWLjVBvVchpQ by outfrost@mastodon.social
1 likes, 0 repeats
@squish @lumi you don't need scientific research to condemn genai on moral grounds, but a more complete picture certainly helps see why it's literally demon tech
(DIR) Post #B4TrbVCGrY0IEHGbbc by walnut@thesoftestpaws.net
1 likes, 1 repeats
@lumiugh, lost my first draft because my computer froze. Fortunately, I was able to take a photo of most of it.For me the entire technology just misses the point. The main reason to use it for writing, digital art, music, programming, etc. is because you don't care about the process, only the output. But we already have enough to read, look at, listen to, and I'd argue, code to last several lifetimes. The only people who care about churning out "content" are capitalists who don't care about the art or what form it is. And if all you care about is the volume of content, we already have enough.There are many reasons to write something of your own besides just the output. You may enjoy writing for its own sake, or it helps organize your thoughts, you want to make something new*, or have your own voice heard. Using an LLM to churn out slop doesn't help any of those. There's also an element of "if a human didn't care enough to write this, why would I care enough to read it."Some with any other art, like playing music. Someone can already listen to any recording of Bach by the best players in the world, or their favorite band, but they still learn to play it themselves, or attend live shows, or write their own music. You can see that tech bros just don't understand the point by insisting that nobody actually enjoys learning an instrument.Heck, this is part of why people care about sports too I guess. Nobody (very few) watches sports only for the world records; they watch what's live now even if Michael Jordan is retired and his entire career recorded. They haven't yet figured out how to try to replace football with a bot, but I'm sure they would if they could.LLMs also are not able to average out something into the best. And if you want to make the best (which is not the only reason to make something), you cannot learn without doing, so using an LLM to do it for you will hurt your efforts in learning.This last part is my worry with programming especially. It's not controversial to say the state of software was pretty bad before LLMs, but adding more code of mediocre quality that people definitely always review the output of (they don't. even if they may intend to) is not going to help things. I'd much prefer to stop the treadmill and actually try improving the quality of software, tooling, documentation, accessibility, security, etc. than using a tool that drowns us in a volume of slop that's impossible to review, made worse by nobody learning any expertise.*For non fiction this can include reporting on new events, novel research, or a new analysis. LLMs aren't any good for that. For fiction new doesn't mean "100% original", but that doesn't mean the average of all previous works is new or interesting either.
(DIR) Post #B4Trj8Ycv3w0KaMdcW by lumi@snug.moe
0 likes, 0 repeats
@walnut i feel much the same way here, it strips the humanity out of everything and just turns everything into pointless slopthanks for writing this :neocat_heart:
(DIR) Post #B4UDjzB5TaQGWPAClk by monkee@other.li
1 likes, 1 repeats
@lumi@snug.moe Article i just found brings up some interesting points.The gen AI Kool-Aid tastes like eugenicsGhost in the Machine director Valerie Veatch wants you to understand how race science has shaped this moment in tech.https://www.theverge.com/entertainment/897923/ghost-in-the-machine-valerie-veatch-interview
(DIR) Post #B4UDkqAYSIUKs6K1nU by lumi@snug.moe
0 likes, 0 repeats
@monkee thank you! :3
(DIR) Post #B4UEeDE8QeBGJLOrUu by monkee@other.li
1 likes, 0 repeats
@lumi@snug.moe When you download A movie cause you broke but want entertainment you are a criminal pirate. :thepiratebay: If you download EVERY Movie to feed the slop machine you are an inspiration, a visionary and very important for the economy. ✨Funny how that works :feelslainman:
(DIR) Post #B4UEheAGjQeIj2EHXk by lumi@snug.moe
0 likes, 0 repeats
@monkee the law has never applied to the wealthy elite. those parasites always find ways around itthey try to distort the whole world to their whims, and it's disgusting
(DIR) Post #B4UEyYzvCAfZciSKYK by monkee@other.li
1 likes, 0 repeats
@lumi@snug.moe I would even say law was originally created to protect wealth - rights for everyone came much later...
(DIR) Post #B4UF87KytHGsjSrVZI by snugglybun@void.lgbt
1 likes, 1 repeats
@lumi this is more of an anecdote than any "hard" evidence but:genai is one of the reasons i struggle with making anything related to art, like music or drawings or poems or whatever... like why even bother if i'm gonna be replaced by a soulless machine, and then most people won't notice anyway...i mean i know that's more or less me spiraling, and most people will notice how the LLM will never be able to replicate the qualities that make myself my own being, also even if genai is here to stay (sadly) we can still fight and rebel against that shit in the spaces we're on.oh and also, the rise of the genai bubble in many companies coincided with them abandoning any pretense, either being genuine or not, of caring about showing and pushing progressive values and instead worshiping the orange man, the muskrat and other right wing shitheads like peter thiel. like it turns out that siding with a corporate machine that disregards any human rights is against any image of being "woke", no matter how you try to put it. and of course corporations are ultimately capitalist shitheels who focus on what can make them the money the most, and since now the moneymaker seems to be being a rightwing shitheel and the ai bubble.... aaaaaaaaaaaaaaaaa
(DIR) Post #B4UFGsCaX6COOg6yNE by lumi@snug.moe
0 likes, 0 repeats
@snugglybun anecdotes are fine, i feel people's opinions are also valuable heregenai does have that chilling effect on many people, which is yet another way it is stripping humanity away from everything. it's, as freya holmér says, a parasitic canceryeah, corporations were never on our side, we can only temporarily use them to get some ends met, but we can never ever rely on them
(DIR) Post #B4UFRjMnirFkBF48EC by snugglybun@void.lgbt
1 likes, 0 repeats
@lumi yeah...corporations are people... evil, twisted, abusive, exploitative, kyriarchic, capitalist, etc people who we should never allow to have the power they have on our lives rn
(DIR) Post #B4UFRjbgpVA8vQ21R2 by lumi@snug.moe
0 likes, 0 repeats
@snugglybun i think corporations are on a class of their own, one an individual person can never attain
(DIR) Post #B4UFUd8rK53Nk7vXUG by snugglybun@void.lgbt
1 likes, 0 repeats
@lumi sorry i was trying to reference that one mitt romney quote and mock it to the ground lol
(DIR) Post #B4UFVGBi9PVYqGYrqa by lumi@snug.moe
0 likes, 0 repeats
@snugglybun oh lol, gotcha :3
(DIR) Post #B4UGWj66Fe7ykNCRuq by walnut@thesoftestpaws.net
1 likes, 0 repeats
@snugglybun @lumiwhy even bother if i'm gonna be replaced by a soulless machine, and then most people won't notice anywayI have half baked thoughts of encouragement if you'd like to hear them :3
(DIR) Post #B4Uf87PbD9wPltgQTI by oblomov@sociale.network
1 likes, 0 repeats
@lumi may be relevant (written a couple of years ago):https://wok.oblomov.eu/tecnologia/continuous-content-generation/
(DIR) Post #B4UiSf2KCmQPi7Hf72 by ledoian@pleroma.ledoian.cz
1 likes, 1 repeats
@lumi For me, the part of discussion on training data seems to often forget the vetting of the data. The “garbage in” part, where the internet was just shoved wholesale to train on, without vetting whether the stuff to mimic is actually what should be mimicked. This has resulted in bad behaviour (as seen in Tay.ai), discrimination or code vulnerabilities.Links:https://hackernoon.com/your-ai-coding-assistant-is-probably-writing-vulnerabilities-heres-how-to-catch-themhttps://www.forbes.com/sites/jodiecook/2026/03/20/vibe-coding-has-a-massive-security-problem/https://www.veracode.com/blog/ai-generated-code-security-risks/https://www.reuters.com/article/world/insight-amazon-scraps-secret-ai-recruiting-tool-that-showed-bias-against-women-idUSKCN1MK0AG/https://psaif.org/2024/10/18/microsoft-tay-chatbot/https://ficpi.org/blog/impact-ai-gender-discriminationhttps://www.ohsrep.org.au/ilo_report_generative_ai_and_gender_inequality_sn_798(I also remember reading lately about an experiment where genai was given a description of a situation and was asked to argument which of the described people was smuggling drugs – and it switched the reasoning just depending on the race or gender. But I am unable to find it now.)
(DIR) Post #B4UiTmPQHOtqrjcaC8 by TheSunnyOne@eldritch.cafe
1 likes, 1 repeats
@lumi if it helps at all, we've been slowly putting together a playlist of videos from various creators about AI.Generally covering the environmental, moral, ethical, social, mental, etc, impact of AI:https://youtube.com/playlist?list=PLkAYSkf6JX4UKiImB-IlXBW2f7aPN5f8G&si=Fw0mis9yO_ObQksE
(DIR) Post #B4UoAz1Vksm8D9rClc by ledoian@pleroma.ledoian.cz
1 likes, 1 repeats
@lumi My another point is against code generation specifically: All other issues with genai aside, it is harder to read code than to write it (from own mental model of what should be done). Therefore I feel like either generated production code is not as carefully created, or code generating AI actually decreases productivity (as more code can be written than carefully reviewed after being generated).This is more of a philosophical point, so I am not aware of sources, but at least the premise is mentioned in:https://www.coderabbit.ai/blog/its-harder-to-read-code-than-to-write-it-especially-when-ai-writes-it (uh, the irony when someone finds the same problem and then “solves” it with another layer of genai, with a similar quality unassurance considerations…)https://medium.com/@FlowStack/reading-code-is-harder-than-writing-it-heres-how-i-finally-got-better-at-it-aa1c14ffc1d8https://www.joelonsoftware.com/2000/04/06/things-you-should-never-do-part-i/
(DIR) Post #B4UoBuOWS03C923t32 by ledoian@pleroma.ledoian.cz
1 likes, 1 repeats
@lumi I also think the generality of some LLMs is bad. At least the general ones (like ChatGPT) were (to my knowledge) trained to “produce human-like output”. Most of the time that is not what they are used for, though (like advice or coding help). However, being “human-like”, people are susceptible to believing the output. (It will even apologise for being wrong! That does not make it right, though, just more humane.)Basically, voluntary delusion from a thing that is a “trained” manipulator and psychopath. I don't think that people really want/should use that.I've written a similar thing about that here: https://pleroma.ledoian.cz/notice/B40GVCaiLOrgh97r96