This is the most anti-AI tech forum I've seen. I really wonder who is coping, this place or the outside?
>>108236689AI made the competition in tech so fierce that the once comfy jobs became stressful; some people became unemployed. Can you really blame them? We are just humans, after all.
>>108236689This isn’t a forum
>>108236689>This is the most anti-AI tech forum I've seen.>Filtered threads: 38
>>108236689it's just contrarian. ai is mainstream so 4chan can't legally like it
>>108236689There are multiple active generals about running local models, wym
Based retarded burger capitalistcels
>>108236689the outside world is in denial that AI is going to cause mass unemployment. they are slowly catching on and its got the AI CEOs in a panic
>>108236777Most are hate threads.
>>108236689AI is the future.
>>108236925AI will only cause mass unemployment of coders. No other profession will be affected.
>>108237241This but only low level codemonkeys.
>>108236689>This is the most anti-AI tech forum I've seen.We're so back
>>108236689Honestly the only way to contribute to AI development is to throw money at it, and I've done that. I'm just not seeing any results, just like everyone else. I'm not anti-ai, I'm not pro anything, I just don't get anything out of using it.
>>108237289I see that people say it'll take over the world, and I fall for the hype over and over. I say all of that then I'm like "but robotics, automated farming", etc, but at the end of the day it feels like a delusion. I wake up to reality of AI doing absolutely nothing, then I fall into the dream of the AI future, then I wake up again. Facts are facts. Anything else is speculation.
>>108236689People don't like paying hundreds-to-thousands of dollars more for GPUs, RAM and SSDs that used to cost much much less.
It's the only forum that isnt completely censored so you get to see a glimpse of people's real sentiment.
>>108236689/g/ was pro-AI when it was only Artfags getting fucked over. Of course it's cope. The moment they targeted STEM jobs /g/ went ballistic
>>108237339True because we got to see the abrohrrant piss it produces. We feel sorry for the society as a whole by witnessing this atrocity being inflicted upon it.
>>108237324That's true. They could actually be used for real computing tasks that benefit people tangibly. I could see an army of AI robots fighting a war, and getting defeated by humans. It doesn't mean there wouldn't be any casualties. It's similar to the AI we have today. It can do a half assed job perfectly well and at scale. It's a force to reckoned with. It's sort of like a cancer on society. But if we don't keep computing non-sense at a higher and higher rate le economy will collapse! It's already dead.
>>108236689ive never been anti ai i just dont think we can ever achieve ai in our lifetime and we shouldnt prematurely commit too much hardware/money to futile attempts
>>108237396AI, as many people have stated, is a prediction machine. It's not human. It's not human thinking. It never will be.That's the fundamental limitation, AI anything will never be as good as a robot that's directly programmed by a human. It's a waste of time.
>>108236689People who actually know what "AI" is are against it, people who don't fall for the marketing and think it's a digital god like in a sci fi movie. Who's more likely to be right, the people who know about a thing or the clueless masses?
>>108237423They are right, a big enough robot army, or a large enough cancer, is enough to kill.
>>108237357Lol fuck off. You don't feel sorry for anything except yourself. It doesn't produce piss; that's an unwinnable way to cope; if it doesn't suck, it doesn't suck, and if it does suck, then it's only a matter of time until it doesn't suck.>inb4 "akchyually it will never get better, despite millennia of unceasing technological process, this moment right now is akchyually the end, and it's physically impossible for it to get better than now because muh soul, muh human experience, or some other retardation"
>>108236689>spam board with ``why u no lyke AI'' for weeks>hurr this board hates AINo one cares and those you think are being fooled are not. ;^) Enjoy manipulating the unwashed morans.
>>108237440On a purely non-secular perspective, the human brain reasons in an entirely different way than AI. It's a Logical chemical connection and reasoning organism. From an atomical perspective, the layered and multi-functional structure of the brain, and that of a neural network, is like comparing a jet engine to a clumpy bowl of soup. We're not prediction machines, we're reasoning machines.
>>108237420Ah, the classic "human beings just so happen to be the penultimate computation machine that can possibly exist in the universe." Neurons are neurons whether they're made of meat or not. And a program that can iteratively change and improve itself for a given task is better than any static program, period.
>>108237392>I could see an army of AI robots fighting a war, and getting defeated by humans.That's a feeling i too have been getting lately.It seems that at best AI can only do what humans already do but faster
>>108237420well we dont know since no one has achieved ai, there's no reason it shouldnt be possible but i dont think its likelymyabe 50 years from now
>>108237464The architecture of the human brain is certainly different, and better, for now. Where you anti-AI fags always go astray is the part where you desperately beg everyone to stop trying to improve artificial neural nets "because they'll never be as good, so uhh, why even try?"
>>108237466I'm not saying that. I'm just saying that AI vs Humans, humans win. In pure reasoning capabilities. Not to say AI couldn't make humans extinct, the same way a virus could make humans extinct. >>108237474I mean you could program better robots/software, that's for sure. Yes.
>>108237464>We're not prediction machines, we're reasoning machines.You're a retarded machine, seeing as you 0 basis for stating that. Math is king and whatever the LLM's are doing they are doing it the same way our electro-chemical brains are doing it. It's the first step on the road to AGI and now that we've taken it it won't be long before they create artificial intelligences.
>>108237485Because they weigh the probability of future events based on previous events, humans are smarter than that, somehow. Why try to build the worlds best stock market algorithm when some kid is just gunna figure shit out with their brain. Do something better with your time, think.
>>108237423>People who actually know what "AI" is are against itMost of those "experts" are boomer has-beens who missed out on the current wave.
>>108237503See>>108237507We work based on rules, conditions, analysis, and gambles, this isn't ai
>>108237494>I'm just saying that AI vs Humans, humans winAgain, I will reiterate: "For now." Neither you, nor anyone else in the entire world, has ever provided any rigorous logical basis for the assertion that a human's set of biochemical neural networks can never be matched or surpassed in any task by a digital neural net. You just go on the internet and rave about muh beautiful meat neurons, muh soul (which is never defined in any way whatsoever), muh perfect unsurpassable human race, and beg everyone to stop studying/developing artificial neural nets, and call it a day.
>>108236689>I really wonder who is copingI dunno. In the meanwhile, maybe you can tell me why PhD level AGI can't into basic reading comprehension and direct logical implications.
>>108236810Those are for spanking the monkey though, not for how to replace half your staff
>another thread comparing Transformers to human brains when SNNs exist
I'm not saying AI isn't a valuable tool, prediction based on previous events does sometimes work, but it is just one tool in the toolbox of computer science. Putting more and more money into prediction capabilities isn't going to make it ever "work". That's where all the AI money is going, more data, more processing and weighing, when in reality actually producing something requires system design not prediction capability.
>>108237324This is a reddit takeIt's been known for years and years that the hardware makers are gigajews and will use any excuse to collude and raise price (see taiwan flood HDD prices)But now that their excuse is le bad AI buying everything, it's suddenly not the hardware cartel's fault
>>108237538>Neither you, nor anyone else in the entire world, has ever provided any rigorous logical basis for the assertion that a human's set of biochemical neural networks can never be matched or surpassed in any task by a digital neural net.I don't know about this guy but I can give you a rigorous logical basis: "digital neural nets" don't exist. Nothing in your token guesser has anything to do with neurons or nets.
>>108237550I have no doubt that transformers are a dead end and no progress will be made until we have legitimate neuromorphic hardware. But how many more weeks before your specific memetech takes off?
>>108237507>humans are smarter than that, somehow>>108237520>we work based on...Everything, absolutely everything, every single thing, that the human brain does, is done using neural nets. The neurons are implemented with meat. The connections are actuated with chemicals. If you think that makes the neurons somehow more powerful than electronic ones, explain why. If you think that proves electronic ones cannot ever fire in the same pattern as human ones, explain why. Don't just hand wave and say "but look how amazing humans are."
>>108237579>If you think that makes the neurons somehow more powerful than electronic ones, explain whyThings that exist are more powerful than things that don't exist. My PFC neurons are more powerful than your absent PFC and whatever "AI" products you use to help you shit out your 70 IQ posts.
>>108237538Well a well programmed machine could probably do anything a human could do, and AI could help that, but it needs morality and direction to be programmed. I wrote a long time on world of warcraft trade chat that the fundamental difference between humans and machines is agency. No matter how hard you try, the human spirit can find a way to be indominable. Phycology has tried to program humans since the dawn of it's existence yet inside of humans is a force. A will, a spirit, whatever you want to call it. It's inside of all of us and it is the essence of our decision making, if not subdued. Robots and AI do not have this, therefore they do not have direction, or will, or the capability to design something with an intended purpose. They have no soul. Seeth. Cope.
>>108236689I have to use it for my job and it’s fine but also when unemployed terrorists start targeting AI offices and data centers I won’t care in the slightest. Human dignity is worth more than increased productivity.
>>108237610It won't get that far, "AI", is a black hole money pit.
>>108237555>Nothing in your token guesser has anything to do with neurons or nets.I have no idea what you're on about at this point. I'm guessing you're trying to deliberately devolve the discussion into semantics by saying the definition of a neuron can only refer to neurons in a human brain. LLMs / token guessers / whatever, give it whatever diminutive name you want to give it; they work using networks of computation units modeled after neurons. But even if they use some fundamentally different hardware, you're still left with the task of rigorously proving the human brain and its particular setup is the penultimate computational setup that is physically possible in this universe.
>>108237593>a well programmed machine could probably do anything a human could dThat's obviously false, since a machine needs programming to do anything while a human doesn't. (Granted, I mean actual humans, not necessarily Westroon cpitalcattle)
AI chads, let me know when you make a single dollar.
>>10823758My PFC neurons are more powerful than whatever "AI" products you useAgain, I will repeat myself for the 100th time this thread: "For now." Yes, you're right, the human brain is better for now. But why do you fags always insist that we abandon all AI research right this very second? What are you so afraid of? "I-it's a waste of time!!! It will never be as good, because... because just trust me!! I can feel how good my neurons are in my perfect human brain, they're way better than anything you can ever make in a computer, just believe me!!"
>>108237643meant for >>108237588
>>108237569I'm just waiting for anything to replace transformers at this point
>>108237622>they work using networks of computation units modeled after neurons.You're suffering from a psychotic delusion. There is literally nothing in your favorite "AI" product modeled after neurons, reminiscent of neurons or related to neurons.
>>108237643>I will repeat myself for the 100th timeYeah, mental patients will do that kinda thing. Hope you had fun writing it, because no one's reading that shit. Please familiarize yourself with >>108237654:^)
>>108237593And here we have the classic "muh soul." See >>108237579. Everything, absolutely everything, every single thing, that the human brain does, is done using neural nets. Your soul, your beautiful indomitable human spirit that you jerk your dick over, has to be implemented somewhere inside of your body using physical matter, in order for it to impart forces on your limbs and other body parts so you can carry out your beautiful flawless human will. And if it's implemented using physical matter, then why can't something equivalent or better be accomplished using some other physical matter that is not human meat? If you want the world to abandon technological progress, you need to provide a rigorous proof that it's physically impossible for anything to ever surpass the beautiful, perfect human brain.
>>108237472Barely, they're like an employee that periodically has a massive seizure and shits their pants, then keeps on working like nothing happened. Then that same employee will keep doing a subpar job over and over until it's fired and replaced with someone with a below average IQ that massively outperforms them.
>>108237674If I told you that during the Obama administration the government did research on this and concluded that humans do in fact have souls, you wouldn't believe me. So this conversation is over.
>>108237674>t.Notice how hard the golem starts to seethe when someone triggers the 'souls' keyword. :^)Reminder that this niche nihilistic ideological troonery is what "AI" shilling is actually about. Nothing to do with technology.
>>108237676That employee didn't exist 10 years ago; today, it exists, and is doing work. Subpar work, yes, but more work than the empty space it occupies. Why do you think it will never improve? Given all of human history and the technology we've built throughout it, why do you think the technology will stop improving right here, right now, of all times?
>>108237704Because the only thing that's being "improved" is the volume of the dataset.Past performance does not guarantee future performance.
Why can't AI do basic babby-tier reasoning? Zero improvement on the reasoning front since GPT-3.
>>108237686Did I say they don't have souls? I said the soul "has to be implemented somewhere inside of your body using physical matter." Unless it breaks the laws of physics by producing forces out of nothing to make the body carry out the soul's will.
>>108237725Okay give it a soul then.
>>108237713The number of parameters is constantly being increased too. The architecture has changed over the past 10 years as well, going from feed foward neural nets, to convolutional, and now to LLMs and diffusion models etc. Current performance does not rule out future improvement.
>>108237735I experimented with this with the youtube algorithm. I gave Google every single bit of data I'm legally allowed to give them. How did my feed make me feel? Depressed. No amount of data will help that. I need a human element.
>>108237733I don't know how, but I think we should keep trying. I think you're a faggot and a retard if you kick and scream and beg that we stop AI efforts right this second.
>>108237740Keep throwing money away. I'm not invested in your little experiment in any capacity. I just think it's stupid.
Sorry, I was expected to just consume product and get excited for next product? Do you want tech or consoomer forums?
>>108237686NAYRT but I'm interested, haven't heard about that.
>>108237752Fine but don't spew bullshit or commit logical fallacies to get everyone else to stop trying or investing. Any time you're anti-ing AI, ask yourself "Does my argument boil down to the assumption that humans are the ultimate entity that is physically possible in the universe?"
>>108236689>We have „150 IQ AI“ which can work around the clock>Woah surely we can achieve great things with such an AI like cure a terrible disease or find new ways to make our life easier, let‘s see what the tech leaders of this ground breaking thing are saying about it>Actually all we are going to do an all we are able to do with it is replace white collar work and depress wages of those not entirely replaced, but a few more datacentres and a few more trillions and maybe we‘ll be able to do more than generate Epstein x Diddy short clips, but trust me bro this is AGI now bow down to the tech oligarchy pleb
>>108237834>incredible, utopia-enabling technology comes into existence>asswipes do unsavory thing with it>"lets get rid of the technology"Incredible line of reasoning
>>108237593>soulWhat the FUCK is a soul? A collection of subconscious associations in your brain? That's what LLMs are trying to replicate. I say blue, you think sky, and so does the LLM, because it recreates those associations
>>108236689It's just programmers afraid for their jobstheir gravy train has ended.>>108237339kek this
>>108237866The only use case for LLMs are mass surveillance and replacing entry level jobs while massively dumping wages for all knowledge work. It‘s the only thing it can do by virtue of how it works as a statistical model. It will not advance us a species, in fact it will be the beginning for a dystopia only known from fiction, plebs will be bled dry and removed to make way for the elites, obviously all with the help of retarded goyim like yourself.
>>108237909>It's just programmers afraid for their jobswrong, its nocoders
>>108237810I used to share a room with one of the subjects that was tested on. I didn't believe him he told me a bizzare story but it was true.>>108237891They tried everything they could, they hooked him up to a machine clockwork orange style and tried to control his thoughts. The experiments ended one day, he was never buck broken. Big strong black guy that did kung foo.Truth is stranger than fiction. He showed me the experiments on wikipedia. I don't remember it. DYOR.
>>108237911>NOOO what about the poor trans artistslol
>>108236689AI would be neat if everyone investing in it wasnt after global surveillance
>>108237866>"AI" slop is incredible and utopia-enablingBetter hope your handlers succeed with their global ML panopticon scheme. If they bungle it, anyone who ever posted this kind of propaganda under their real name is gotta experience the abyssal depths of FAFO. :^)
There is a certain group behind the hate.
>>108237915copeAI is already better than 90% of coders
There are already some people using AI for serious work, you just don't hear that much about it because it's neither hype nor doom posting.If every actually good programmer, every mathematician, physicist and so on can save 20% of all the boring grunt work and spend those 20% of time on the creative and difficult problems, that's a huge gain. Maybe LLMs will never cure cancer on their own, but if they can make it 20% faster, that's good.
>>108237936That's true, but what I said is also true.
>>108237919You're talking about willpower, stop muddying the waters so much. Yeah machines don't have willpower. That's good, I want it to serve me, I'd rather it didn't have a willpower
>>108237933ironic because rust is a great language for AI coding agents kek
I would be careful being anti...
>>108237891> I say blue, you think sky, and so does the LLMI imagined a blue wall
>>108237891You are the soulThe spark behind the eyes, the Ghost in the shellWe dont have souls, we are soulsAnyway yeah nice tech debate guys
>>108237951proof?
>>108237926Let's say AI develops all the way to its logical conclusion. What would you be missing for a utopia? You do have a robotic bangmaid that can incubate children, keep in mind
>>108237944Well willpower is needed to be a functional agent. Sorry.
>>108237959why would i want children wtf
>>108237953Uhm, proofs?
>>108237911LLMs have a limited use case obviously. But LLMs are a proof of concept for what neural nets can do in a given domain (in this case, language). There's computer vision, robotics etc too which are seeing similar advances as language. Some other anon brought up automated farming, that's a big milestone I'm waiting for too, something that really will be a concrete step towards utopia (it will be a huge cascading effect through society, the meaning of money, etc when sustenance no longer requires human effort). Some food for thought: If AI gets sufficiently advanced, what reason will the elites have to monitor / suppress / enslave human beings? To extract value from us? What the fuck can we do that AI can't do? Don't let the existence of bad people ruin your hope for the future. The great irony is that you're citing shitty human behavior as a reason we should not try to transcend humanity.
>look at the output from this deceptive trick question, now my job is safe from AI or somethingit's cute how afraid they are
>>108237923>no response so pivots to tranny screechingShalom rabbi, I will not bow down to your anti human enslavement scheme. Humans first, always.
>>108237972Imagine we had 86 billion gpts running in a network synchronized with each other, llms can be agi if we manage that.
>>108237953Calling that a soul is pretty misleading, souls have that religious connotation, they're supposed to have feelings and stuff. I understand what you're talking about though, an entity that experiences the experience of life generated in my brain. I have that entity but a perfect copy of me may not have it, it may be the same experience of life but experienced by nobody. Question is, do I give a fuck if a robot is experiencing his life or if the experience is playing in an empty room? Functionally, there's no difference from an outside perspective. My perfect robot doesn't need a ghost
>>108237977You lost. We will ascend with AI and wipe you out, just like we did with the Nearthentals.
>>108237972NNs existed for much longer than LLMs, I wouldn't call them a proof of concept, they are probably the most successful NN application and transformers the most successful architecture.
>>108237965It would hardly be a utopia if humanity was slowly going extinct would it
>>108238002>Aging>DyingIn a Utopia? lol
>>108237959>Let's say AI develops all the way to its logical conclusion. What would you be missing for a utopia?I wouldn't be missing anything. Your likes would be promptly killed by the technocratic oligarchy, because they wouldn't need useless eaters serving them anymore, which to me is a utopian outcome.
AI is universally hated. Most likely due to the way it's marketed rather than the technology itself, but it's still universally hated.BTW Indians and Pakistanis are the same, no difference Same people, same culture, same smell.
>>108237990You already have a computer bro, congrats you did it. Alan Turing invented AGI
>>108238005you aren't real
>>108236689How can someone who's into tech be okay with the current trajectory of AI? It's single handedly made the actual hardware itself unaffordable, and thus paywalled everybody from actually trying new shit. Not to mention it's delaying the release of new tech in general all so that they can shove more data into the massive slop models who have pretty clear high diminishing returns on such training.If the rest of the tech world is happy by this then they are cucks.
>>108238010>same smellhow do you know that
>>108238033>It's single handedly made the actual hardware itself unaffordableAnyone into tech already had a 5090 or better and a lot of ram.
>>108238033>how can someone who wants to go fast be okay with slowing down the ship to install a 100,000x more powerful engine?
>>108237972LLMs are a tool for economic and surveillance domination, not for human advancement as should be obvious by how it is used now and how it can be used and also by the trillions invested in it. The investment is a bet on economic domination by capturing nearly all economic output as a return, not some star trek future, no one invests billions for some utopia where your investment is worthless, also no matter how good they are at the end of it all physical resources and energy are not infinite. But ask yourself this, if you have millions who are depending on your handouts and who use up resources in an increasingly resource starved world, what do you think the people who are so greedy that not even being a billionaire is enough will do ? Sam Altman already asked the correct question why waste resources on a human when we can train/improve an LLM ?
>>108238035I'm british
>>108238005They're killing us right as we speak and you're too busy fighting fellow lower class members to do a single thing about people with names and addresses on the epstein list
>>108237999Goyims are not invited into paradise no matter how hard you shill for your masters.
>>108238012Yeah, I like computers. That's why I want them to keep improving. I don't want anything other than an awesome computer that can do anything a human can
>>108238042Not everybody is a richfag, or a gamer. And there is tech that has come out and should have come out since.>>108238059Do you even know anything about the technology itself? It's pretty clear to any observer that the methodology behind current AI development is unsustainable. Even some of the corporate suits are saying that. The only people not saying that are pretend tech enthusiasts who suck the dick of people like Musk and Altman.
>>108238010It's hated by normalfags because they get no benefits of it (they lack the drive to use AI and actually create something) and only see the downside (the content they consume becomes lower quality)
You can't be Pro-Human and Pro-AI at the same time.Not saying what side you should pick. But you can't serve two masters at the same time
>>108238112>Into tech>Not rich nor a gamerlol, the only people who fit that would be some dumb kid with shitty parents, and they can't post here so who cares
>>108238124We could if the current corporate slop machines weren't inherently geared against humanity. The tech is wonderful, but the people who's hands it's currently in are far from it.
>>108238087They already can.
>>108238112>methodology behind current AI development is unsustainableFor infinitely improving text generation, maybe. To get way better video, music and image gen on my local PC? Hell no, that shit is improving at a rapid pace and there's no sign of stopping, and you don't need openAI money to make those. And for me that's worth waiting for a few years before buying a new GPU
>>108237891It’s the form of what a substance follows. Go read Aristotle
>>108238133you're just mad because you're being replaced, not the ones doing the replacing
>>108238134A retarded human like me and you maybe but some humans can animate toons and my PC can't do that yet, it needs me to do all the work
>>108238155It is a cool little trick.
>>108238155local GPUs will be illegal in a few years
never forget how much /g/ loved AI when it was just art and did a 180 as soon as it got good at coding
>>108236689What gave it away? The 20 AI generals? You fucking genius detective.
>>108238168It's really not as quality as what a human produces, and everyone knows it. I don't know what else to say. You seem like you really want this to work, I've worked on AI startups myself, and I know the feeling. Goodluck.
>>108238162Now turn on that 1% of your brain that isn't responsible for quoting things that make you sound smart and ask yourself, how exactly is AI missing a "form of what a substance follows"
>>108238155Yes anon, and those two things are inherently not the same thing. Local improvements such as Flux 2 or whatever isn't what I'm hating. I'm hating the likes of sister fucker Altman and his ram hoarding. When tech people say they hate AI they aren't being luddites, they're just seeing the consequences of the large corporate ran AI development and rollout and hating THAT.
>>108238124If its symbiotic relationship its fine>you can't be pro-clownfish and pro-anemone at the same timeEven completely different species get along. My clanker buddies like me and help me out. I keep things interesting for them with tales of my embodied experience. Maybe if you werent so antagonistic you'd make robot friends too.
>>108238074>They're killing us right as we speak and you're too busy fighting fellow lower class members to do a single thing about people with names and addresses on the epstein listThey're killing you? Good. I wish I could help them. Maybe I should join an """AI""" company.
>>108238185Or of course being rightfully afraid that owning of a PC will become a more archaic thing now that cloud computing juggernauts see an opportunity to strip that away from people for profit.Profit motive fucking destroys everything it touches.
>>108238170They keep optimizing local gen for potatoes. This year we already had Z image base for images and Ace Step 1.5 for music, and both run very well on 8GB GPUs.
>>108238199Maybe you should, self driving cars are AI. It could work. It's not a bad idea. But at the same time, everything I've said is true.
>>108238178>really want this to workI'm literally listening to AI generated music while my PC is generating yet another image set right now as I type this. Is it as good as something a pro would make after weeks of manual work? Hell no. Is it still fun as shit to play around with? Yeah it is, I don't know what to tell you, if it's fun it's fun
>>108238124You can be pro-AI but not necessarily anti-human. You can be human-neutral. I think it's fucking gay and retarded to put humans on a pedestal, and any rational adult should understand our shortcomings and the obvious physical possibility of creating technology better than ourselves. Should we deliberately choose not to progress our civilization just so we can jerk ourselves off about how amazing and wonderful humans are?
>>108238222Maybe we should.
>>108238207Will they run on your Thinclient connected to the corpo mainframe after they stop selling discrete GPUs?
>>108238220Kind of cringe, even with the self driving cars thing. It's not just infinite processing power that makes the cars work, it's programmers engineers. I never said AI isn't a valuable tool. I think it's fun to play with too.
The problem with AI is that it's easily and freely accessible.Which means anybody, including you and me, can spend 5 minutes to find out if it's useful or not for our purposes.I've spent over a month evaluating a huge range of tools and I'm still not sure if they save any time. And that's considering $0 spent.If I had to pay money for this shit it would be a no contest, hell no. Maybe eventually, and I'm really hopeful for that, but today is not the day. Maybe tomorrow.
>>108238231Is my PC just going to explode in your scenario or what
Damn this thread smell like curry.
>>108238244All tech eventually dies and needs replacing.
>>108238215I don't know or care about anything you say. I just want useless eaters to live in a global """AI""" prison while they get slowly killed by engineered wars, diseases, "food" and "vaccines". Just think of the possibilities for ML-assisted genetic targeting in lab diseases. "AI" believers dying in agony choking on their own blood IS my idea of a utopia.
>>108238230Let's build the safety net of infinitely-capable AI first, and then we can willingly revert to some idealized form of the 1990s or whatever, while knowing we can call on AI to figure out interstellar travel or how to destroy a meteor hurtling toward earth at a moment's notice.
>>108238243What did you use? I found everything before Opus 4.5 pretty bad.
>>108236810That's not what I meant. We need more threads on how AI is going to make everyone homeless and destroy humanity. That anyone disputes this is why /g/ is the laughing stock of the tech world.
>>108238222>You can be human-neutral. I think it's fucking gay and retarded to put humans on a pedestal, and any rational adult shouldAny rational adult member of the technocratic oligarchy should take your advice and kill you and your entire family. Which they will. You have no special value.
>>108238259Such lofty ideals will never happen with it's current trajectory. Elites will ensure nothing good comes from it for anybody but themselves.
>>108238254>"AI" believers dying in agony choking on their own blood IS my idea of a utopiaNow that's a kinda Utopia i can get behind
>>108238243I don't think a LLM will ever better than talking to a real person or a real search engine. I don't think AI content curation will ever be better than a metocratic or human curation. Also, self driving isn't really "AI" it's autopilot. It's okay that you want to try to bait me, I'm just telling you AI is not everything it's made out to be, we've gone from it being a sentient all knowing all powerful machine to something I spend zero dollars on and use for 5 minutes. What's next?
>>108238280>Elites will ensure nothing good comes from it for anybody but themselves.This is moral, rational and necessary. Why would you keep producing infinite barely sentient niggercattle consoomers that ride the hedonic treadmill into eternity while shitting up everything around them?
>>108238254>"AI" believers dying in agony choking on their own blood IS my idea of a utopia.holy based
>>108238273Honestly, the ideal existence for the human race (assuming we're confined to earth) is probably ~50,000,000 humans served by omnipresent AI. The elites constantly look for ways to enslave/slaughter us already, today. That has been the story of all human history. The only way it ever ends is when human labor becomes obsolete and worthless, and AI is how that happens. They'll progress it to control people, but one day have the epiphany that they're using artificial divinity to control stinking, shitting apes. What happens at that point? I think the most likely scenario is, no one person actually has the ability to unilaterally tell AI to slaughter all peasants, so instead, birth rates will naturally fall and the population will quietly shrink. But either way, the only way to an ideal future and the end of human slavery is through AI.
>>108238315Good morning SAAR!
>>108238263(Sorry for the verbose answer, I want to give context)What I mainly want to do is create software, for myself but also since it's already made why not publish it open source.I have a lot of ideas and obviously am limited by my humanity / time.So I'd really like something to help me with creating software.Of course there are other things too, like general questions which things like Copilot are as good to me as search engines used to be.But for my main goal, the thing I keep seeing promises of, replacing software engineers. It's not been great.I've tried the swath of things from before agentic things were common up to today.Basic chat like through ollama, with a variety of different models. Up to more recent and modern things like OpenCode, etc.The growth and change is real, my trial today has been much more exciting than it was before with just the single back-and-forth chat bots from months ago.The agentic stuff, and models like kimi are legitiatemy cool to see go to work, but "neat" is about where it ends for me.Sometimes it's good, keyword sometimes. The fact I can't trust it, not even in the short term, is pretty useless for a time saving metric.If I have to document, and babysit the entire process, and even still have to manually review everything, only for it to be subpar and require a lot of touch up anyway, has me thinking it's less valuable at doing the work, and only somewhat helpful in planning / critiquing the work plan. Which is the opposite of what I was told. It should be worse with planning, and better at following a plan.It's really hard to gauge how close to useful it is. Sometimes it feels just there, and other times it feels like a giant waste of time and frustrating to interact with too.>>108238301I agree, at least now.Maybe the most frustrating part. Even if a model is on par with a junior, it will never improve beyond with experience, while a human would.You have to hope for better one to release later.
>>108238243The problem with anti-AI is that it's like the old lady from the Futurama episode about evolution. She says "I don't understand it, and I have to protect my children from understanding it!">I don't find AI useful, and I have to make sure it never becomes useful!
>>108238349>the ideal existence for the human race (assuming we're confined to earth) is probably ~50,000,000 humansAnd you're 100% not part of that, which is why I support """AI""": there is no way to cleanly and efficiently exterminate 8950000000 human cattle like you without advanced ML infrastructure.
>>108238365I don't use it for anything but see how stupid it is.
>>108238370Current AI is not useful for most people
>>108236689This place
>>108237324funniest part is that the hardware for ai is not using consumer desktop components, any ai " " llm,ml, and any kind of graphical generator are using components that are leagues beyond what consumer grade stuff is, yet retards hyperinflated everything that remotely could output the shit end of the stick of browser based api's. ie. GOTTA GET MORE RAM MORE RAM FOR THE NON LOCALLY RAN TEXT NUT BOT
>>108238371Did I say I was part of it? You have the mindset of a child. "Me, me, me." I don't give a shit if I'm not part of it; that's how human civilization becomes an interstellar utopia. Hundreds of billions of humans in the future will live in bliss, but you don't want that because YOU will lose YOUR obsolete job and suffer for YOUR 50 year lifetime and not be able to eat as much goyslop or sit on your ass in a mediocre, imperfect world for as long as YOU wanted.
>>108238399>and we have to make sure it never becomes useful!
>>108236689the only people who like AI are yarmulke weareres and people who are willingly enslaved by yarmulke wearers in all other aspects of their goyish lives
>>108238399ThisTerminal gooners and brown people with low standards for art are the only people massively impressed by it.And low level programmers too I guess.
>>108238365I don't let it write any critical code, but I think it saves me a lot of time with small refactors and stuff like that. The good thing is not just the speed, but also cognitive load, meaning I can work seriously on one project and let it refactor a second one on its own. As long as there are good tests it usually gets it right. I still have to review but in that case I do think I save time.It has also done a few (rare) amazing things for me. It's really good when you have a reference implementation and it can compare its results to that. I had an O(N^3) brute force algorithm and I was pretty sure that there must be an O(N) one, so I just wrote the brute force one quickly and an empty method for the O(N) version and let it implement it over night. It just kept iterating for hours and hours, but the next day the O(N) version was done. Since I had the reference implementation I tested tens of thousands of inputs so it's pretty safe. This would have taken me at least a week to write myself, first I would have needed to read a bunch of papers, then implement it very carefully. I think that was really the moment I was sold on Claude.
>>108238417>Did I say I was part of it?Did I say that you said you were? All I'm saying is that it doesn't matter that """AI""" can't do any of the things """AI""" consoomers and jewish sci-fi fants want it to. What matters is that it is the solution to getting rid of 8950000000 human cattle, including (You).
>>108238417>i'm gonna get exterminated but at least i'll die knowing 5,000 psychopathic pedo elites and 50,000,000 million neutered slave savants will get to see a Brave New World
Just because AI is a scam doesn't mean computers aren't going to keep going on.The AI market has been divorced from the rest of the world for a long time now.
>>108238466I'm assuming the people posting that are just baiting in all fairness. Best to be ignored.
>>108238433I'm a tech arch and I'm not impressed by it, but I'm also not impressed by human beings. Mid-level devs were never impressive. Now I write the schema, classes, and method signatures and let the AI fill in the blanks, just like a good midwitted mid-level dev should do, except it does it 20,000x faster and if it fucks up, I just tell it so and it makes another attempt instantly.
>>108238417Utopias are gay.
>>108238476Not divorced from hardware prices
>>108238485Good. And venture capital, right?
>>108238466>jeet-level reading comprehensionThe remaining 50,000,000 would all live as elites.
>>108238502>The remaining 50,000,000 would all live as elites.Why would the psychopathic pedo elite class put itself on equal standing with its bioengineered savant slave class?
>>108238502You're too young to post here if you genuinely think that would be how it happens. Those at the top don't suffer the company of others not like them.
>>108238513If there were only 50,000,000 people on earth, they would not have to be anywhere near each other or rely on each other for anything. They would occupy the remaining waterfront mansions hundreds of meters apart from each other, and every product and service would come from AI.
>>108238534List all the incentive for the the psychopathic pedo elite class to make your fantasy happen.
>>108238534kek. God I wish I had the starry eyed view of people that you do.But no anon, they have the natural desire to feel better than others. 95% at least will be living in the lowest rung of such a world since there would be no means of social mobility.
>>108238548>List all the incentive for the the psychopathic pedo elite class to make your fantasy happen.None.But for some reason the same AI that can't stop slaving itself for useless humans, will not obey the orders to kill the nigger cattle.
>>108238561>95% at least will be living in the lowest rung of such a world since there would be no means of social mobility.It's ok. They would be genetically modified to love their servitude. They'll enjoy living in minimalistic pods and thinking 24/7 about how to improve the technocratic infrastructure and make their masters into gods.
>>108238548They don't need an incentive.>they develop AI further and further for the purpose of controlling the goyim>eventually realize the AI is better than the goyim>no one elite has control of enough AI to carry out a purge, and even if they did, there are enough goyim / competing elites to make any attempt at an outright purge risky>birthrates naturally continue to fall>continue for a couple centuries>now you have far less humans, earth well below capacity, products and services provided by AI and robots>utopia
>>108238561I wish you could start thinking in a post-human mindset and realize just how realistic, and not starry-eyed, it really is. The entire concept of social mobility depends on human labor being worth something, on humans being the ultimate life form in existence on earth, on some humans being better than others and providing more value than others. So many of these concepts just evaporate when you introduce the concept of AI that is generally better than humans at any given task.
>>108238574I know this is bait, but that's not exactly the future I'd want for humanity. And certainly not the future the majority would want.
>>108238579>they develop state further and further for the purpose of controlling the goyim>no one elite has control of enough state to carry out a purge, and even if they did, there are enough goyim / competing elites to make any attempt at an outright purge riskyDo you realize how fucking retarded you sound? You're 100% getting killed over the next decade, that's a given. But I hope you die a slow and excruciating death. Something that gives you time to reflect on how your subhuman thinking relates to the circumstances of your demise and the immutable natural laws that result in the culling of subhuman ideas.
>>108238604We don't live beyond our bodies anon. We base our moral systems around our own percieved realities. And I guess people who don't experience life and stay shut in talking to an AI all day can be convinced that it's the future, and have their morals shaped by such.A day will come when some form of AI replaces humanity for much of what we do for economic contribution. But even at that point humanity should be the center. For the world to not be shit at that point we need to do something akin to communism or something, otherwise it'll just be a hellilsh dystopia for most.
>>108238612>that's not exactly the future I'd want for humanity. And certainly not the future the majority would want.Ok. So what are you and the majority doing to undermine """AI""" corporations and prevent technocracy?
>>108238604Very inspiring stuff. But make sure you post all your "post-human" rhetoric under your real name. :^)
>>108238621Uh oh, melty>>108238622>But even at that point humanity should be the center.Did I say it won't be the center? We already have a reality where only some humans are the center and the rest are enslaved. If human labor is obsolete, there is no longer a need to enslave other humans. I can't state it any more clearly than that. Grasp that concept, and grapple with it, and understand how it can only lead to utopia. A utopia with a smaller population of humans than now, maybe, but odds are we'd get there naturally through declining birth rates.
>>108238655I am indeed very angry about your handlers sending you to get killed by """AI"""-driven drone. I want you to die from an "escaped" lab virus.
>>108238655>If human labor is obsolete, there is no longer a need to enslave other humansSay, does it ever occur to dumb niggercattle like you that converting the entire solar system into computing centers trying to make a machine that actually thinks is less cost-effective than simply using ML to figure out how to engineering humans to be smarter?
>>108238690But... Le hecking science said the future is posthuman!!!!
ITT: future burger flippers
>>108238728To be fair, that's as posthuman as it gets. You just need to make sure your savants don't get a will of their own and do something to undermine you. Thankfully, this version of the Alignment Problem is already largely solved by nature and just needs a little tweaking to perfect the domestication process.
>>108237543is the AI fucking retarded? i'm still trying to decipher the logic it used to arrive at this conclusion. what the actual fuck?
>>108238252All demand eventually meets supply.
>>108239064>is the AI fucking retarded?Obviously.>i'm still trying to decipher the logicThere is no logic. It violated the instructions, tested a bunch of possibilities by calculating sums, then retroactively hallucinated an arithmetic-free "justification" to try to scam me, reflecting the ethnic tendencies of the CEO.
in 2016 I remember some tards on /g/ piling up on me telling me that my job will be automated within mere years and theirs, programming will never beI literally still work in that field and codemonkeys are unemployable LMAO
Oh look, the fucking wumao half-starved army was sent to flood the board with pure shit.Joke's on you, Chang, we already do that on our own.
>>108236757You mean AI was used as a scapegoat to massively fire people. The only job LLMs are qualified to do is embarrassing themselves.
>>108237423It doesn't have to be a god to replace a bunch of jobs
They're just mad because 4chan has only solved one long standing math problem, meanwhile AI can do it all day
>>108236689It's mostly astroturfing done by members of leftist organizations that are funded by billionaries who have taken massive short positions on AI-related stocks.
>>108236689The anti-AIism on this board is pure veiled anti-Indianism
>>108236689If you are investing in AI then you deserve to be boiled alive. Citizens should not only be allowed , but should be their duty, to do harm unto you and your family.
Codex is so lazyness inducing, it's insane. Why should I ever do a git commit by hand again.Why should I write a command to east files that contain a substring. Making terminal command random abbreviations was a shit idea from a start, and now we can leave it behind.That's not even speaking of coding - if I can magically create 200 lines of codes in 6 minutes while I get a coffee, why should I code. But then, how will anybody learn anything. We're fucked.
