File: 1743647974775.webp (33.23 KB, 600x400, I, Robot.WEBP)

No. 2472431
ITT: Defenders and detractors debate the morality and utility of AI.
Topics of discussion may include but are not limited to:
>What jobs can be replaced by AI?>How can deepfakes be stopped? >Can AI-generated art be copyrighted? >How does AI insert bias? >What data should AI be allowed to collect? >Will AI develop a conscience?>Can AI chatbots replace human companionship?>Will AI destroy us all?Do not post AI-generated art in this thread. /m/ AI containment thread:
>>>/m/443713 No. 2472451
>>2472443If I write the prompts right, I can force an AI to call me out properly for therapy. No sugarcoating, "everything you do is totes
valid!!" or armchair diagnoses like a lot of quack therapists. I'm also using it to polish my prose, it's pretty helpful and has taught me stuff they never really teach you even with something like a creative writing degree.
AI slop just looks ugly as fuck. Overly glossy, weird faces and hands. I don't care if people want to generate dogs flying planes, I've done it too, but after a while it just gets… lame for me? It's just so easy it makes me feel empty inside kek.
Also AI is used to make CP.
No. 2472542
File: 1743659135013.jpg (37.51 KB, 473x572, 1725027564667.jpg)

I'll start a quandary. When is an AI image generation model "ethical enough"? Say someone makes their own model and loras for making personal art, is that equally bad as sloppeddlers stealing artstyles and making clickbait disinfo for a content farm?
Is a model that is ethical to the public eye possible? Can people against AI be satisfied when people using AI try to choose ethical methods? To clarify, an ethical model would be a self-sourced model with their own images, or could easily be sourced from public domain art free of abuse material. There are also many public domain models out that are free of photography/carefully curated to not be trained on children when the model is designed to be NSFW capable. That's not to mention how most publicly ran models are incapable of rendering NSFW as it is. It is intentionally obtuse to label all models as "CP MACHINES" when the most commonly used models like DALL-E and midjourney can't even render NSFW on their own. If you dont know the effort some developers put into keeping their data clean of abuse material, you are likely too ignorant in the development of ethical models to have relevant commentary on AI ethics. citing that one model trained was on CSAM when you can find 10 others that have explicit policies and data to prevent CSAM and deepfake generation shows that people aren't willing to see that efforts to push AI into an ethical direction, and are only interested in statements that imply that the technology itself is implicitly evil and needs to be erased. Which is impossible, so both sides must be willing to make concessions.
Without the question of copyright and CSAM involved when an ethical model is used, AI imagery becomes a debate of artistic integrity and the environment. To address the environmental factor, the level of power consumed by all generative ai is alarming, until you look at the perspective of what the global power expenditure is. Streaming video is estimated to consume ~250 terawatt-hours per year (accounting for about 1% of global electricity use), while chatgpt, the king of ai energy consumption, is estimated to be consuming around 7 terawatt-hours per year (0.028% of the world's electricity use). If you are concerned about the environment, you could do your part by returning to physical media, but those that argue for the sake of the environment won't because of the convenience of streaming. Does this not denote hypocrisy in the environmental argument? To continue on my core point, the power usage of private models is basically irrelevant in comparison to public models, making it the most ethical mode of AI image generation. The usage of a local model takes the power equivalent of using an air conditioner or car for the same amount of time, where is the environmental outrage every time someone mentions using a car or AC? 3d printing a single small print consumes nearly x8 the energy of a local generation of one image, where is the pouring outrage towards 3d printer bros destroying the earth? It indicates the environmental point is a virtual signalling issue that most people fall back on when they can't target the reason why AI makes them upset.
Destroying the art market and job security should be a justified enough reason to be angry, but it reveals another hypocrisy. Artists are scared about their job security, but show minimal awareness that many jobs have had their job security threatened by automation, and is an unfortunate inevitability that will not render artists into irrelevancy, but will trim the fat of the market for artists incapable of competing. It becomes a grey area that invokes unfortunate casualties, because technological advancements are inevitable for every career on earth. This is one of the most valid points of contention, but in the subject of ethical AI, reveals the majority problem is not with the technology itself, but the industry that exploits artists, and the competitive nature of the artist market.
My ending question is, if ai users are willing to make these contentions to appeal to the sensibilities of those that are against it, will they be accepted by those who refuse to engage in good faith? Or will the goalposts continue to shift, revealing that the core issue isn’t just ethics, but an unwillingness to accept technological change? And most importantly, what defines an ethical ai image generation model to you?
No. 2472555
>>2472461the issue with ai images doesn't really stop at art.
I recently saw what was very clearly an ai chart for medical purposes. Unless we put ai images under a filter that separates them from non-ai, this could be disasterous.
No. 2472559
I hate how, now, if you have a decent enough PC and do some research, you can make realistic deepfake CSM. It's making the dead internet theory a reality and these corpo's behind it are not on the side of the little guy. They want many jobs replaced and, since company towns are making a comeback, I see it more likely that we're paid like shit and forced to slave away in a corporate town than a utopia. Wealth distribution will get smaller and smaller. Technological feudalism mixed with ideocracy. These companies pushing AI have in the past hired people to exploit your psychology to keep you hooked on their platforms. Now imagine that with a mix with CCP level of surveillance.
People act like removing the ability to make AI video/photo will stop the good things it does. It won't. Text AI was fine but image and video generation has no benefit and is dangerous. It's retarded how chatgpt and other companies can get away with violating copyright while we can't.
I see the people who dismiss others concerns as retarded bootlickers. Morons who won't realize the problem until it's too late. If you think the lack of privacy in the modern world is disturbing now, it's going to get 10x worse in the future. The only one being somewhat smart about this is the EU. Keep calling people "luddites" when you're forced to be nothing but amoeba for the rich.
"Wetware" and other weird attempts at making AGI are not beneficial for humanity. Brain chips are not beneficial to humanity. People will stand for anything even if it's against themselves.
Art affects culture. Art is so much more than what people believe it to be. We've already seen the start of a corporate dominated culture. Now imagine that but these platforms and companies selling you nothing but artificial content. Music, movies, TV shows, literature, news outlets, fashion. You won't be able to trust anything unless you see it right in front of you.
No. 2472569
File: 1743662530617.jpg (839.47 KB, 1080x2400, 1000001954.jpg)

Sigh…this thread is going to be just like the antifujo vs fujo threads isn't it?
Either ways I like ai especially for things like role-playing. Ai yumes rise the fuck up.
Yes there are bad things about ai but you can say the same thing about ALL TECHNOLOGY and not just ai. With every new thing you have both bad things and new things.
No. 2472582
>>2472575>nowThey have been illiterate for ages kek
>>2472577Yeah. Lolicons constantly say "b-but we're just getting out the urges, at least we're not touching real kids" but now that AI is a thing, real kids are being trained for AI. So… yeah. The best way to get rid of a habit is to not do it.
No. 2472588
>>2472583>>2472572To be fair I clicked on the first random one I saw and screenshotted.
The actual chatbot I am talking to is alot more normal but I ain't gonna share my private convos.
No. 2472611
Pov: it's 2100 and you can finally have sex with Connor from DBH for real.
No. 2472640
>>2472464agreed csam was already a huge problem and now ai only makes it worse it only gives predators freedom to abuse children (thankfully ai csam is illegal now), the other issue for me is deepfakes and the potential of my images being used in porn, if ai was only used in the medical field or the astronomical areas where it was needed i wouldnt mind it but its not being used for that its being used to replace human hobbies and target women and children more, i was already scared enough to be on social media but now ive deleted my personal account even if i wanted to keep in touch with my family but the risk of some moid using my images is too much
and the only one whose in any fatal danger is me if fake porn gets made of me any woman who keeps supporting image generation ai is only supporting moids who want to violate womans and childrens human rights
No. 2472650
I'm a nonnie working in tech and I'm actually horrified about AI. Now, it's extremely efficient in doing calculations that would take way too much time from humans and can't afford calculation errors, that's for sure. But so much of the hype is artificially inflated by the people developing these models and selling them. Investors are being told that they can cut 30-40% of the workforce because AI increases efficiency to the point the remaining people can just take care of everything via prompting, which is in no way true in practice but rich people don't care. The AI craze is going to damage so many industries with mass layoffs and people going jobless because some out of touch investors thought that a LLM can fully replace human cognition and because they don't appreciate human effort at all, and it's depressing. It's such a post capitalism scenario, everyone's learning skills are deteriorating, including mine. I just ask AI for shit now instead of figuring it out for myself. And why do I do it? Because I'm being forced to. I'm forced to work more efficiently and harder, and I can't meet those demands without asking AI to do part of my work for me. I'm still burned out, stressed, just as overworked as ever, but now there's an added layer of laying my trust on a machine the way I never had to before. I seriously dread for the future. I feel like there's no going back from this and we're done for. Smug people still have around two years to cope with how "it's just natural selection, you just need to swim or sink" because they too, will be replaced soon enough. Why are we as humans so eager to destroy ourselves? UBI will never be a thing, tech billionaires would never allow it despite having enough money to last them a billion lifetimes. Once AI takes over in full, most of us will go to peasant level poverty and misery working 15-hour shifts at massive warehouses to satisfy the 1%'s frivolous life style.
No. 2472680
>>2472650>Smug people still have around two years to cope with how "it's just natural selection, you just need to swim or sink" because they too, will be replaced soon enough. Why are we as humans so eager to destroy ourselves? UBI will never be a thing
>UBI will never be a thing, tech billionaires would never allow it despite having enough money to last them a billion lifetimes. Once AI takes over in full, most of us will go to peasant level poverty and misery working 15-hour shifts at massive warehouses to satisfy the 1%'s frivolous life style.Exactly what I've been thinking, new age feudalism. Pro Ai retards don't want to see it but this is going to kill us. don't want to think of consequences they just want to be lazy sacks of shit. They want their coom and memes. The devil is real and the devil is selfish researchers and the silicon valley.
>"NOO DON'T REGULATE IT, I HOWPE ALL OF YOU ARTISTS ARE REPLACED BECUASE I NEED MY COOM AND MY MEMES. COPYWRITE FAGS AWRE NO FWUN."I have seen so much of this retarded shit on the other farms it's insane.
>Eat bugs>live in shitty section tier 8 housing>social credit system>watch nothing but artificially created entertainment that serves the ideals of corporations and the 1%.>Listen to artificially created music that also serves your corpo overlords.The more I think about it we're going to probaby turn the whole world into some north Korean level shit. We memed ourselves into 1984.
No. 2472714
>>2472443Same
To me AI has been useful for two things. First, Therapy or really just talking about what's bothering me which to me is the equivalent of journaling. This isn't me isolating myself from my friends. I tried but unfortunately they start doing this whole "actually i have it worse than you" dance which is insensitive. I went to a therapist anyway but booking the session takes forever. The second being job applications (interview prep, cover letters and CVs) because that shit is soul sucking anyway.
I tried using it for shopping it's not that good but better than google which pulls all these SEO optimised pages. Which I know Chatgpt also does but from experience it brings about new results. Adding "reddit" or going on tiktok for "user generated content" is probably the best you can do but people get memed into parroting the same response.
>>2472561That's true. Google images is practically unusable imo. At least with Pinterest you can game the algo to avoid or minimise the amount of AI pics you're seeing.
No. 2472734
Ai haters: I hate ai, ai has no value blah blah blah
Meanwhile: this masterpiece
Checkmate ai haters.
No. 2472737
File: 1743682450809.gif (8.82 KB, 150x129, ida.gif)

>>2472461Samefagging
>>2472700 because AYRT is irking me. Like I said, you are describing post modernism and post modernism is dry rot to all creative spaces and all non-retards reject it and its ideology, which only leads to despair, detachment, nihilism and suicidality. If you knew anything about art movements you would not say what you said.
>inb4 muh accessibilityKnowing about art cost me nothing more than the energy to live as someone with a half working brain and the ability to see, and it didn't even take that much time. I am a retard poorfag with no resources and somehow I magically have learned all this.
I don't care about generative AI imitating art despite being a drawfag, I care about actual real world harm it can and will and does cause, I also care about people acting haughty about art while pulling shallow ignorant gotchas. "AI art is good acktually because banana on wall acktually urinal that everyone hated except people huffing farts and students" there are worse post modernist "artworks" like a fucking scrunched piece of paper or a gross retarded porn tape and guess what! Made by the same people who love scammy shit like NFTs and are probably creaming their pants at the thought of putting less effort into their works via AI. If you look at any other art movement it was always about exploring the human psyche and/or pushing their art to be technically better or more experimental and/or communicating the cultural and religious values at the time (and being historically valuable first hand, secondary or tertiary sources while doing so). Do not taint art with comparing millennials of history to that shit stain of a movement that encourages promortalism-esque values and hyper-consumption bullshit. Before you bring up furry and coomer shit, those aren't art movements the same way cartoons, comic books, commercial advertising art and pin ups aren't movements either. To reiterate I am less offended by AI images than I am about ignorance, AIfags ought to set a better example if they don't want to be perceived as hopelessly vapid, make better and more informed arguments about art and I'd have zero reason to sperg. Tell why it's good without comparing it to the bottom rung of all art movements because you do such disservice to your points.
No. 2472787
File: 1743686432655.png (250.8 KB, 512x384, 1703482153132.png)

>>2472774Just like I mentioned here
>>>/ot/2470206Art commissioning has to be one of the worst routes to pick as a main source of income. Not just because of how unreliable it is even if you have the skill, but because artfags are on the front lines to that fear of robots replacing us in jobs. Funny how all the moms who hammered in "become a nurse or a doctor" to their kids somewhat ended up being right.
No. 2472803
>>2472787Art and writing was seen as the safest from AI which is why chatgpt and co are concerning in the first place. No one expected code to replicate what artists do.
“Don’t do art” parents are right for other reasons. Art in general is not in high demand in the first place and you don’t get paid a lot.
No this doesn’t mean art is useless but you have to find different avenues where your skills are needed, which is heavily discouraged in online art communities
No. 2472804
>>2472801Moms hammered it because being a doctor is a prestigious job that pays well in most countries. If it pays like shit it’s still a well respected job and med school is known for being difficult. No one hammers “you have to be a consultant at big 4” for a reason and you only hear about it when you’re an adult
I can’t say the same for artists
No. 2472846
File: 1743690320681.png (175.98 KB, 751x858, zl2hthq.png)

No. 2472857
>>2472848>have fun with your irl scroteYour first mistake is seeking any sort of emotional support from males.
A good female friendship is a beautiful and spiritually fulfiling thing nona.
No. 2472862
File: 1743690927743.png (1.76 MB, 1024x1024, axl animu.png)

AI will automate knowledge economy jobs, white collar jobs. Anything to do with the mind, memorization, or the voice. It will also unfortunately replace all art forms.
The safest jobs will be trade jobs, they still havent made a robot that will fix your plumbing or cut your hair.
I dont think we will ever reach what is called AGI and especially ASI
But a lot of people are going to lose their jobs
if you have some sort of art degree like graphic design, you best start preparing for the worst.
No. 2472900
>>2472848bitch be so fucking for real right now, journals have existed since egyptian nonas were making papyrus or some shit you're not special for not having people emotionally there - not to mention that you yourself say you are venting to a chatbot, at that point you use ai because you know that it can't talk back at you
literally go on a walk and talk to yourself or journal like the rest of us instead of relying on asspats from an algorithm that will coddle you
No. 2472920
>>2472912thank you, if thinking out loud is schizo I'd much rather be one than melt my brain with AI.
When chatgpt and dall-e and co came out I was not a huge fan since I draw as a hobby buy I was pretty neutral, but over time I've found that the people who are fans are consistently not only stupid in their arguments but also smug about it which pisses me off because these are the same people who are supposedly fans of logic and debating.
It seems that we've been slowly spiraling into a learned helplessness and
victim complex as a society, and having a text prompt that you expect not only thinks for you but also listens unconditionally (and takes verbal abuse if you look up any articles about moids harassing chatbots in lieu of women) is only making things worse. I'm not saying we should be insensitive and tell everyone to toughen up, but sometimes shit sucks and you have to deal with it so it sucks less.
>>2472917all I'm saying is that I wanted dogshit art I'd much rather make it myself, it's not like AI art is magically better because mine is dogshit. Of course some people will ree out of fear of irrelevancy but more often than not the people who claim AI art is soooo much better supposedly act like they are responsible for it.
>Baby you were irrelevant before ai and will continue to be irrelevantThis applies to you too "baby"
No. 2472937
File: 1743694521456.jpg (35.02 KB, 735x676, 1710745888450.jpg)

Pro ai fags are very alike handmaiden, when troons took over handmaidens and pickmes kept convincing others it was innocent and it was only a way for "misunderstood" people freedom to Express their "true" seleves only for it to end in total destruction of women's rights and freedom some of those consequences still felt today. Its saddening how far women will fall just to appease the common moid even in turn solidifying their misery and slavery
No. 2473076
>>2472734lyrics:
crocodildo penisini
peperoni masturbini
trippi troppa crocodini
dildo dildo dildo dini
dildo dini
dildo dildo
crocodildo penisini
peperoni masturbini
trippi troppa
crocodini
dildo dildo
dildo dini
wow, a tear ran down my leg. This is beautiful. Thank you ai for writing such a masterpiece.
No. 2473131
>>2473046Everything construction
Plumbing
Carpentry
Electrician
Pest Control
Landscaping
Cleaners
Cooks
Waiters
Sculptors
Firemen
Police
Hairstylists
Manicurists
Miners
No. 2473157
>>2473131Seeing a list like this is pretty grim. A lot of these professions regularly expose you to materials that are
toxic with long term exposure. If they don’t then they destroy your body and you’ll be in pain every waking moment for the rest of your life, often with few to no benefits so you’re unable to even afford medical care. I left a job that is AI safe but physically demanding so I had some quality of life with my health. It feels like there is no good end given the lack of worker protections and rights, especially if you’re a burger.
No. 2473188
>>2473149I used to joke that whenever I'd be going through shit I'd go and take sheep to pasture/have a sheep farm. I have never had contact with that lifestyle but if everything is fucked I might as well give that a go. Work for sheep's cheese with my nigel or something. More realistically though I'd want to learn how to be an electrician or a plumber. People will need electricity to run AI and they won't start to shit outside or something.
>>2473157Wholeheartedly agree with you, we are cooked. Not a burger but in my country most young people were pushed into either becoming an engineer, work as a programmer or become a doctor so practical skills (being handy if you will) is a dying art. I hope that you won't get in a situation where you have to sacrifice your health to make money to pay the bills.
>>2473167people did that with google before though, people have avoided thinking for themselves for a while.
No. 2473937
>>2473933Really? What are you doing where AI is remotely useful for maths?
From experience it’s more likely to give incorrect answers
No. 2473939
>>2473828I defended AI use in this thread
I’m sorry but this is dumb. Just buy or borrow a random children’s story
No. 2473943
File: 1743743478441.jpeg (136.39 KB, 1170x637, IMG_4767.jpeg)

>>2473933I doubt that very much.
No. 2473953
>>2473943I havent had this problem. I use it to check if my homework is done correctly and so far it hasnt failed me. If i had the money to waste on tutors i wouldnt be in college in the first place, so its the best i can afford.
>>2473937I ask it to give me excercises and then i check if i did them correctly.
No. 2474222
File: 1743770246049.png (296.03 KB, 605x637, dbvZy0n.png)

the fact that I legitimately can't tell if this is ai or not makes me angry
No. 2474226
File: 1743770343161.png (546.85 KB, 726x979, 92D31129-1BB2-4D38-A311-E2B836…)

Maybe AI is a force for good after all
No. 2474233
>>2474222The quality on the photo from 1980 is too good to convince me it's not AI. Casual photos from that era aren't nearly so sharp and I can't imagine calling in a professional for 'big cauliflower'.
Also why would a photo from 1980 be in black and white? It's not like we didn't have color film then.
No. 2475535
File: 1743833677485.jpg (1.62 MB, 6259x5147, 1719288659747355.jpg)

AI art specifically is such a big consoom hobby that I'm convinced it got to this big of a debate simply for being spearheaded in the US. I can enjoy AI for a quick glance but it's so hard for me to engage with it because there's no actual thought behind it. I can't study the folds because the AI is copying off of thousands of images of folds in photographs, but doesn't understand the material and gravity, because it isn't human. I can't study lighting because AI does not understand the concept of light, it doesn't know what consistent lighting could be. I can't study color because there is zero choices or reasons why AI chose that specific color palette. I don't judge ai consoomers but the only appeal it has is surface level consoom. Majority of consumers don't want to be given a product that nobody wanted to even put time into it's presentation, but most of the time they have no choice because companies don't care about how they're perceived, they care about profit. Most AI shilling is really pushed by a small portion of the population and tons of higher ups.
>What about jobs? AI is killing writing, drawing, and code!Being a burger I stopped worrying about what I was majoring in because after seeing nearly every industry be gutted right before my eyes I realized that no matter how hard people try to play the system, it's really not their choices that lead to success, it's about luck in a system that hasn't spotted you trying to cheat it. The nursing industry is facing over-saturation hence why nurses are paid in peanuts, business is next on the chopping block, programmers have been experiencing mass layoffs, government employees are hopping back and forth between employment and unemployment. We live in a clown world. I can think of tons of ways AI can take over the "safe" jobs, because no job is really safe when it can become obsolete based off a few rich guys' drunken brunch ramblings.
>AI art is totes better than your average chicken scratch beginner artistI don't care. I want to know why a human decided to draw that, I want to know why they chose those colors, I want to know how their skill level tries to execute ideas they have in their head. I want to know how long it took for them to draw their chicken scratch, I want to try to find the anime or art they were inspired by. I know there's the meme about sovl but sure, call it that, but I prefer shit human art above good ai images because I know there's someone who has years of life to try to express rather than a machine that burns down an entire forest to make a low-tier reddit meme.
>Humans are still picking and presenting the images!If I asked the average ai image user the reason why they chose the image vs. why an artist decided to use those colors, those brushes, that shading style, that technique, that formatting, that expression, that pose, etc; you can clearly see that there's more conscious decision making.
>>2472737 Based. Ai art shillers and anons who doompost about ai looking more aesthetically pleasing only focus on its aesthetic appeal because that's all they prioritize in art. It's never about the person who made it and their decisions in creating it. Peak consoom behavior.
>>2472956 This is a fine compromise! Although I don't think this would ever work because that would mean tech ceos would need to actually invest in hiring new people to curate old material to feed the machines.
Unrelated: There was a study on ai vs real human poetry, and the average normie who does not read poetry actually prefers the ai generated poetry emulating the style of famous poets. However, when the normie was told one of the poems featured was ai, they had an averse reaction to it. People don't prefer ai, but I feel like they do want to learn more intellectually stimulating things. The disconnect between "high culture" things and media made for the contemporary person is because of a lack of content to help transition to "high culture". I hope there are more opportunities in the future to fulfill the niche.
No. 2475700
File: 1743849232431.png (379.04 KB, 640x636, IMG_4773.png)

No. 2476349
>>2472862Being a career artist was always difficult.
If you don't innovate and bring something new to the table then yeah you're fucked. But if you innovate you'll be just fine since AI isn't and can never be innovative
No. 2476794
File: 1743906270663.webp (25.56 KB, 640x504, IMG_5476.webp)

>>2476715nta, I want nonna to expand too. I took a couple art history classes in college and enjoyed them, but I have 0 exposure to anything like that in my working adult life. If nonna has any recommendations on study materials (books, even YouTube channels) that do a good job teaching normies about fine art, I would love to hear. I want accessible but not dumbed down to the extent that something like a pop sci TV documentary is.
No. 2477212
File: 1743947561958.jpeg (521.17 KB, 1440x1920, IMG_5503.jpeg)

>>2473943I’m a rock climber and noticed nothing wrong with this at first kek
No. 2477489
File: 1743958315591.png (144.95 KB, 268x350, 3423423423.png)

When humanlike robots with artificial intelligence come out, I want to marry one.
No. 2477539
>>2472862>AI will automate knowledge economy jobs, white collar jobs. Anything to do with the mind, memorization, or the voice.It depends on the protections the give it. Just because certain jobs can be replaced, doesn't mean they will, or else retail would've been automated almost completely a long time ago. Majority of jobs people already have are essentially "useless".
The human populace needs busy work so they don't have time to become independent thinkers.
No. 2477959
I'm temporarily working as an AI trainer (mostly on 'expert' level tasks and fact checking) and it's insane and appalling to me that high school and college students (and everyone else but it's especially scary in educational contexts) unironically use LLMs for 'research' and assignments, or use it instead of google/actual real sources of information. I spend hours every week correcting AI models that fuck up the most basic shit, hallucinate constantly, give completely nonsensical answers to simple questions and can't follow simple instructions so I have no idea what 'research value' AI can possibly have beyond asking it for a nearby restaurant or something. Even though Google's search algorithm prioritizes a bunch of for-profit/propaganda websites and often leads users to wrong and adulterated information, at least when you use Google you need to actually synthesize info from multiple sources and at least you know what the sources of the information are, but asking a chat model for 'research' it will just spit out some words it strung together and you will have no way of knowing where it got that information from, let alone whether the information is remotely correct. It's more often than not entirely made up. Not only is this terrible because it's teaching people to be unable to think for themselves but it's also teaching them to just believe words spat out at them by an algorithm. It's like asking a schizo rambling on the street to lecture you on physics or European history. Why even go to school at all if you aren't trying to learn the most basic life skills like reading, writing, looking up and synthesizing/sourcing information, basic logic/critical thinking, arithmetic, etc? Why pay to go to college especially if you're not going to actually do your own homework and try to learn something?
The thing with using AI for 'research' is it not only can't tell you what its sources are, but in most cases it can't even access/read webpages or texts. It has some (usually somewhat out-of-date) training data, which from my understanding largely consists of reddit posts, and if it's a paid/sophisticated model can sometimes 'web crawl' in the sense of knowing the titles and the first couple sentences of a given webpage, but it's not going to be able to actually find and read sources for you if they weren't part of the training data (and what is in the training data is usually a black box). I know a lot of people are more concerned with what AI will do to art or how it will 'take over jobs' but I'm more concerned that AI use will just completely kill what is left of education entirely. I was teaching college classes right before the boom of people using chatGPT for homework and the level of writing, reasoning and basic math of the average student was already appalling even though they had to sort of do their own homework back then (most of them plagiarized). I can't even imagine how retarded the average college graduate is going to be now that most of them don't even attempt to do any of their own learning. AI can't possibly replace a lot of the key work that people do at the vast majority of jobs, so replacing the workforce seems like it's not going to happen, but it can replace people's critical thinking capacities with a puddle of goop and it already seems to be doing that, so how are those people even going to do jobs in the future?
No. 2477991
File: 1743984851949.jpeg (26.55 KB, 750x152, image0.jpeg)

Idk about you but geepeetee knows that they look at your post history, I will not reveal the method on how I got it to prove this to me. Run while you can
No. 2478078
>>2476715>>2476794Here is the article I'm referencing:
https://www.nature.com/articles/s41598-024-76900-1What I mean is that people do not prefer AI because if you ask the average person I think they see AI as a quick tool but definitely not an avenue for genuine artistic expression. The only exception being the occasional person who wants quick gratification and a pretty image with no substance. I also think the average person wants to learn more intellectually stimulating things through the rise of podcasts, video essays, and the constant circle jerk of "intellectual" discussions on twitter, despite all of these avenues being extremely poor in research and information quality. In my anecdotal experience, normies tend to see "high culture" (fancy art and academia) as inaccessible, which it is. Academia and art tend to be in this incestuous circle jerk of trying to stay inaccessible to maintain value within the communities it resides in, when there is a large benefit of sharing these subjects to a broader audience. Majority of media that appeals to normies tends to dumb itself down by corporate executives, keeping the media that is more intellectually stimulating with more opportunities to analyze and dissect in this realm of "fancy people with time and money to do fancy things." people get scared of seeming stupid and ignorant when they shouldn't be. There is an untapped/sparse market that features films, shows, illustrations, and poetry in a way that appeals to a broader audience, but also in a way that gets these said people interested in more advanced media.
> I took a couple art history classes in college and enjoyed them, but I have 0 exposure to anything like that in my working adult life.This makes me so sad! I live near a city so it's very easy for me to hop on the metro and go to a local museum, and I know that's a rare opportunity most don't have. If you check out the art book section of a book store there are plenty of books dedicated to a variety of tastes that explain in depth techniques and historical contexts behind a lot of iconic art. There's also pretty pictures and it looks nice on a coffee table so there's the added benefit kek. If you can find local art shows you can not only enrich the community you live in but also have a deeper connection with the art from where you live. Having conversations with art curators there can seem awkward but they're really excited to talk about the artists featured.
If there's any anons ITT that are interested in poetry or philosophy and want something to get your foot in the door I recommend Sisyphus 55 on youtube for philosophy, and Roughest Drafts on youtube for poetry.
I can talk about how AI is a symptom of a bigger problem but I don't want to sperg too much kek. Something something AI and the loss of community and the rise work culture something something.
No. 2478118
File: 1743991738536.png (127.2 KB, 1270x454, geminifaq.png)

>>2478079I'm under an NDA so I technically am not allowed to say but not to my knowledge? They don't actually tell us explicitly which model we're working with most of the time (even if it can be pretty obvious) and they use codenames for all the projects anyway. I know that I've trained a variety of different AIs with (supposedly) a variety of sophistication levels and they all suck and hallucinate nonsense constantly, though.
Unrelated to my job specifically but you can also just look up sources on how LLMs work and they'll openly admit they just pull sources out of thin air. If you go to the google Gemini FAQ website for example they straight up admit that when Gemini gives you sources, they might just be 'similar sources' (aka random unrelated webpages) to the ones it actually used (which makes sense considering how LLMs actually work) unless it's directly blockquoting. And it's supposed to be one of the more cutting-edge ones for 'research' apparently. If you were an undergraduate student writing a paper for your Subject X 101 class you would have to actually identify the specific source you used to back up each piece of information - even mainstream shitty journalism articles often do this or hyperlink to where they got the information. You can actually follow the sources back and see where the info is from if you're interested. The idea of LLM as a 'research aid' which spits random unverified information at you and then gives you sources that are 'similar' to what it might have used is a travesty and basically ensures that anyone using it for research can't confirm any of the information they're getting is true or even sourced from anywhere. It's just some bits of words pieced together by an algorithm that 'sound right.'
To give one irl example of the pure retardation of the Google AI assist, I was looking something up about ancient Chinese history and the AI assist spit out a whole detailed multi-paragraph essay about how it was common for women in the Tang Dynasty to take multiple male concubines lmao.
No. 2478360
>>2478078NTA but this is interesting and I think it does partially explain why so many young people are drawn to AI and stupid social media shit like video essays and retarded podcasts, especially zoomers who lost years of normal socialization and schooling because of COVID lockdowns (which I think drove people online, permanently, more than they otherwise would have been and also made a lot of young people skip the developmental phase where they learn to instinctively distinguish between reliable and unreliable people/information and to debate/argue points rather than just passively consuming them - if you were in a physical classroom at school you could have class discussions about a topic with your friends and teachers in real life but in zoom school or doing some dumb online assignments you're basically just passively interacting with information and other people through screens).
>I also think the average person wants to learn more intellectually stimulating things through the rise of podcasts, video essays, and the constant circle jerk of "intellectual" discussions on twitter, despite all of these avenues being extremely poor in research and information quality. There is a huge issue with people, especially younger people who aren't used to doing research the old-fashioned way, physically visiting museums, etc. feeling like academic topics and high art are too 'inaccessible' to them. It's not just because those things are inaccessible like you said in your post (although I agree about the academic and artistic circlejerk you're describing) but also because they don't have the tools to access them and they haven't been taught how in the modern education system. And just our culture generally has deprioritized third spaces, going out to physically interact with art, etc. But people still want to feel knowledgeable and educated and shit like podcasts and video-essays feels like an easy entry into understanding complex topics even though it usually isn't and is usually incredibly dumbed-down and ill-researched. Not just because of corporate media but also because of influencer culture and 'content creation.' Youtube incentivizes long videos but young video-essayists don't have 40min worth of shit to say about a topic so they'll just mumble about vaguely related topics for most of that time and people who aren't accustomed to getting info from good sources 'feel' like they learned something even if they actually didn't. I know I'm repeating a lot of what you said but I don't think it's always deliberately done by corporations or academics trying to stay relevant and esoteric, it also just happens because of how people are interacting with media now.
I don't know if this is what you were getting at but I think the whole shitty podcast/video-essayist model of infosharing has actually primed people to use AI in a way that feels unthinkable to people 10 years older because those of us who are used to pre-content-farm times can sense how fake and surface-level AI 'art' and 'research' actually is, while kids used to just passively consuming 'content'/wordvomit online from their peers can't really accurately sense that because they haven't experienced better alternatives. Even like 15 years ago the process of looking something up on the internet you were interested in was an active process that involved a lot of thinking about HOW you were going to search for it (what keywords to use, which hyperlinks to follow, what sources were reliable, etc) and it also involved a lot of reading and even asking real people around you what you should be looking for. I remember when I was in middle/high school I would ask teachers and random older people for book and art recs before I went to the internet, but now with chatGPT you can just ask the AI to recommend you something and it obviously won't be personalized or intentional, just based on popularity. You don't have to spend a long time actively searching for resources so the whole search for them doesn't feel intentional or valuable to you. Sure a lot of people still get a vague sense that AI art is garbage, but they often can't identify off the top of their head what the non-garbage alternative is or what about it makes it better. And you end up in this whole online ecosystem of other (mostly young) people like yourself recycling and rehashing information from chatGPT, trending hashtags, trending video-essay topics, etc. The internet used to feel like an endlessly branching tree of resources that could lead you into increasingly niche rabbit holes but now it feels like a funnel that's dumping everyone in the same bland, nowhere place.
No. 2479888
File: 1744125198106.png (71.7 KB, 674x320, ai.png)

No. 2479990
>>2479888that has to be bait
>developing chatbots with the help of another chatbotIm pro AI but this is fucking insane
No. 2481811
File: 1744247832826.png (120.43 KB, 1680x859, sprechengpt.png)

>>2481764this is true. picrel is what it said it would be like as a human, with no input from me (just asking it to choose things like a job, continent, country, city etc that it resonated with)
No. 2482962
>>2482952It's too positive and can reinforce bad/negative behavior that you tell it. It's no different from "actually that's totally normal and
valid" from your average school counsellor plus "you're so self-aware and that's such a good question". Though if you ask it to be mean and properly call you out it can as well, I suppose.
No. 2483023
>>2482999Monday is fun but it still does the positive reinforcement thing no matter what you write (unless you say you're gonna kill yourself etc), just more snarkily
another thing I find annoying about chatGPT Is that it will always take the side of the person who is writing to it. For example, last week I wrote about a situation where a "friend" manipulated me but from "her perspective", as in I was pretending to be her and described what she did and that I reacted poorly, and it saw no issue with her behavior, said it was good and that I must have personal issues that have nothing to do with her, which was nonsense because the manipulation was very straightforward. Then I let it know that I was actually the other person (the one who got upset) and it started taking my side and saying that the "friend" took advantage of me and shit like that lol
It's similar to how normal therapy works where the therapist will never call you out and always support thinking that even if your behavior caused the other person to lose their shit, it's always their personal problem that has nothing to do with you and you should distance yourself because obviously the other person is wrong.
No. 2483032
>>2473131Small women can’t do a lot of these jobs. There are also a lot of women who can’t do physical jobs for various reasons like back issues or injury due to pregnancy and birth. Most trade jobs will be filled up with men because men give other men priority for opportunities. I hate the whole “go into trades” meme because everyone forgets (or doesn’t even consider) that a huge portion of women can’t actually do jobs that require high, or even mild, levels of strength. Also good luck being able to continue doing a highly physical job through menopause or through aging. Women have much higher rates of osteoporosis for example.
Men love AI because they know it won’t fuck them as hard as it’s going to fuck women. They can even keep doing trades into their 60s. They will be receiving most of the benefits from AI including returning to a predominantly male workforce. I say all of this as a technonny myself by the way. People in tech have so much power they will not be stopped unless the entire population decides to switch to flip phones and aggressively boycotts almost all forms of modern entertainment. That will never happen.
No. 2483068
File: 1744336839916.jpg (95.65 KB, 736x736, cedbfe2ac0a6ce648ce7bbfeffe3bc…)

>>2483032I don't think the "go into trades" meme is because men know women "can't do these jobs". Most of the jobs listed actually don't require that much physicality so even smaller women can definitely do a lot of the jobs listed. Of course small/weaker/older/pregnant women will have trouble with the ones that require more strength like construction and mining. That's just true. Even then, a lot of women can still do it if they have the right physicality. only 11% of construction workers are female however women are wayyyy less likely to have injuries/deaths on the job (probably because they actually follow OSHA regulations kek) with female flaggers being an exception. Of course it's fair to argue that men are better suited for these jobs when it comes to physicality, but I really have a better argument as to why women shouldn't fall for the "trades are better" psyop. Men love the trades faggotry because they're mad that women are out numbering them in college, so they're trying to devalue education by arguing that trades are ACKSHUALLY superior. They wanna fill it up and block it from women as soon as they're able to psyop it as "superior". It's also why they want a recession, they want the jobs they deem useless (female) to lose value. There was already a recession like 16 years ago though and it definitely didn't send women back to the kitchen or increase the birth rates kek. So women should keep investing into education, we should be training for leadership positions and not for the mines. Our brains are more valuable than theirs'. It's going to be very funny when in 2050 the average woman is like 15 IQ points over the average male. It's retarded because AI should obviously fill up the most dangerous jobs that give you shit like black lung. Men and women alike, nobody is immune to the long term effects these jobs have on you. Technology shouldn't "replace" anybody but it should definitely be used to prevent the shitty effects that would actually result from manual labor.
>>2478360>>2478078As a socially retarded autist it's kind of weird watching normies be socially retarded. I feel like the internet used to be an escape now I just find myself wanting to spend all of my time outside, off of my phone or computer unless I need to work or go on lolcor. I love walkable cities (please don't call me a gay nerd I will hurt myself), I love sunny days, I absolutely love museums, and I love hanging out (no phones involved). It all feels so purposeful while just scrolling on my phone feels lazy and overwhelming. AI and social media algorithms sort of made me appreciate my life more in a very unintentional way and i do hope more people wake up to it. i literally just can't find myself caring about the latest trend or discourse anymore. it all begins to feel like the same gray mush.
No. 2483073
File: 1744337270136.jpg (52.45 KB, 526x689, 1000066156.jpg)

>>2483023>It's similar to how normal therapy works where the therapist will never call you out and always support thinking that even if your behavior caused the other person to lose their shitThat is not how normal therapy works, several times my therapist has called me out on how I've acted towards other people, and at other times she's stopped me in my self destructive tracks by calling attention to my self harm. Imo that's where AI chatbots fall flat. They provide validation and surface advice and that's it. They can't get down to the nitty gritty, or focus on how to sufficiently live in the present/future, because they don't know how to guide a counseling conversation and their "memories" are too shit to bring up the relevant information from "past sessions".
No. 2483319
Sorry for posting TikToks but it's terrifying that we're already going this route after such a short timeframe of AI getting popular. It's not AI that is necessarily bad but it's just too easy to exploit for people who have no morals. Everytime some techbro tries to downplay people's fear into "hurr they're just afraid of advancing technology", just remember that they try to gaslight you into believing that this is no big deal.
No. 2484033
File: 1744411244791.jpg (255.12 KB, 1146x2048, GoNS1ilWsAAJTLD.jpg)

Jesus.. therapy at number one
No. 2484058
File: 1744412039362.jpg (205.83 KB, 896x896, eagle.jpg)

>>2484050>and they will only affect women.not if we get our most powerful de/g/ens at work
No. 2484065
File: 1744412388597.jpg (22.27 KB, 564x269, 1699233268683.jpg)

>>2484058Oh boy, I just
wonder which certain scrote they're gonna make deepfakes of first. Like gosh, it's just on the tip of my nose.
No. 2486744
>>2484033I'm not really surprised that the top uses are more or less self-improvement, I honestly think it's one of the few good things about AI. Even though you shouldn't but I think people feel comfortable just blurting all their problems to a chatbot. You don't feel like you're getting judged and can elaborate as much as you want. You control the flow of the session and you know, it's free, you're less likely to get ripped off. Like
>>2484269 said, the initial stages to therapy are painful and arbitrary. You need to "click" with a therapist (patient-client relationship) before you feel comfortable to let out the rest.
Goes without saying that if you need medication, chatbots and AI suck but if you just want to talk to someone it's good. It doesn't make therapists obsolete either but will probably cut down significantly esp people that just need to organise their thoughts and can't justify spending a lot of money on that. I don't see it any different than when I used to look up dumb "how to -" when I was younger. Unlike some websites or communities you're always encouraged to improve yourself or reminded that whatever you're feeling is v a l i d. I'll take that over the downward spiral that online communities can create.
Downside based on experience is your conversation skills also deteriorate because you don't reallyyy need to be concise with AI, it can fill in the blanks for you
It's also not a substitute for friendships, I don't think this needs to be stated but just in case. Real social interactions are irreplaceable, even if you're on a discord server with strangers
>>2483068>As a socially retarded autist it's kind of weird watching normies be socially retarded. I feel like the internet used to be an escape now I just find myself wanting to spend all of my time outside,I relate so much to this sentiment. I used to be happy because it made conversations easier because we're all connected and stuff but now everything feels fake. This will sound very terminally online but I hate how I can tell what platforms or even accounts people follow. It used to be good to find like-minded individuals but I hate that now.
No. 2489534
File: 1744830076868.jpg (53.56 KB, 540x405, 1657002895900.jpg)

How fucked do you think accounting is thanks to AI? i want to know if i should drop out or nor
No. 2489581
File: 1744831967036.jpg (208.92 KB, 1024x515, 6121178_original (1).jpg)

>>2483032>men DEI hire each other True
>women can't do these jobs Debatable
No. 2489905
>>2489696Depends, how close are you to graduating?
(Btw, you might also want to ask your profs or other people in the know about this. We're all just anonymous dummies after all)
No. 2497807
>>2490562They need to program ChatGPT to stop using the em dash so we can reclaim it. I dunno why some people lose their shit over the
fact that the em dash is an AI signal though. it shoves multiple into every short response, meanwhile most people are too illiterate to use the em dash and if you’re not using a text editor or mobile keyboard you have to use a shortcut. Unless it was academic writing I always just used -