...they will keep us alive, right?
There will always be a job for (You)>>533741927
>>533774282Just sabotage communication towers and power systems and then go straight to the homes of the rich and powerful
>>533774282>for real workOh, were previous versions not?
Cgpt would t even help me cheat on games via cheat engine because it violates the ToS. Can you imagine the AI throwing roadblocks at you during real work? Like OOPs sorry I can’t scrape this website it is against the TOS heehee. Fucking joke of a platform
>>533774628Have you seen what agentic AI can do? It's absurd. You tell it what you want done and it drives the computer just like the best computer science, physics, engineering, mathematics, medical, chemistry, sociology, psychology, and literally any other field you can name experts would drive it, all combined into one. >hur dur do work with your handsThey are making robots (that can self-repair) do that work too.The age of the fleshbot is over
>>533774798Rules for thee, not for Sam Altman & coYou are literally not needed
When will they be profitable?
>>533774282Can it count R's yet?
Its UBI or mob riots, let's see what they choose
>>533775291Is that what you do for a living? Count Rs?
>>533774282>muh AI>muh UBI>muh kike sóy-fi
MONEY FOR NOTHIN'CHICKS FOR FREE
>>533775362No but I can do it much better than a language model regardless.
>>533774282>guys, this time for real it's happening, it's over, we've created something too powerful. pls send money
>>533775036is that from the gif where the robot is raping the counter top?
>>533774282Every single aspect of the Klaus Schwab NWO video, especially the line "and we've trained you to support us", was just rightoids projecting onto other people. Everyone was just pawns to openly cheer on the construction of the panopticon NWO hellstate
>>533774282This is going to be the next weed legalization topic for the next 60 years
>>533775394LITTLE PEDER
>>533775036start the most successful business on the planet then. 2 week retard
>>533774282>imageHow large portion of its training data is AIslop that they accidentally included? My guess is that we're already nearing the 50% mark
>>533775425How much food will that earn you?>>533775436No, in this one the robot submits itself for automated repair when it realizes it has malfunctioned:https://x.com/bladeofthes/status/2047076046285971733?s=46
>>533775362They market it as an intelligent being so it should be able to do the basic shit that any actual intelligence does.If not, they should stop marketing it as an intelligent being and be open about it being a mere autocorrect on steroids
>thinking the elite wants less peopleno they want more consumers/slaves lol
>>533774282does any still use that shit?
>>533775535Tikkun olam can only exist when everything you see and hear is fake and gay.
>>533775591I get plenty of strawberries.
>>533775596Does it matter what they call it? It can do your job. You aren't needed anymore.
>>533774282AGI in 2 more weeks
>>533775605Contributing to the image limit of a thread negates putting "sage" in the options field. Thought you should know. Thanks for the bump friend :)
>>533775599they used to need slaves because they needed the labor do you think that will still be true
>>533775646That is a lie
>>533774282You're a retarded star trek pipe dreamer if you think they want to do anything with the jobless masses except hunt them for sport.
>>533775714If it can't currently do your specific job today, agentic AI has entered a feedback loop that will keep self-improving until it can. You can keep saying "not my problem" but it is very rapidly going to become your problem which is why you should be thinking about it now.>first they came for the CS majors, but I said nothing for I was not a CS major...
>>533775691Why would I not want my comment bumped? You think I come here to be ignored?I am saying your are trash and telling everyone to come see what a retard you are for using AI.
>>533775936I don't use AI, thoughbeit
>>533774282Surely the billionaire that have made their wealth by walking over the less fortunate will turn into saints once they have the power the annihilate us.
>>533774282If it’s not Mecha Hitler Im not interested
>>533776047The dream of effective altruism was always "allocate as many resources as you can so you can enact the highest good".That's what Sam Bankman-Fried told me, at least...
>>533776041Well…sorry for calling you a retard then.
>>533774282>...they will keep us alive, right?Why?If humans can no longer meaningfully contribute to the GDP, what even is our purpose?I feel like capitalism isnt going to last long if gen ai turns out to be actually useful
>>533775691Not anymore, they changed that years ago
>>533774282They’re basically going to make us all into the irl equivalent of a janny. Sad.
>>533775102>When will they be profitable?They will, anthropic started the enshitification because they don't have enough compute. When they overcharge for a token based service it will be profitable. They just need to keep the current model for a little longer, just long enough to make sure local is dead and buried. Once it's done, it's fuck the customer time and massive profits.
>>533774282>...they will keep us alive, right?Some
>>533776144It's alright my friend.
>>533774282I'm retail wagie. I work 10 hours a day 6 days a week. I stock shelves, talk to customers, do cash register, listen to my boss BS, clean up. I don't require a tech support by 10 pajeets, don't need my battery charging after every hour of work for 5 hours. And I can do all that for $12 dollars and hour. No meme robot will ever replace me.
Everyone who talks about AI doing all these people jobs never seem to articulate beyond very vague things because they're managers, not the people doing the jobs. Every AI advocate sounds like some jeet on linkedIn. I'm a software engineer now - there's nothing I can actually use AI for except as a stackoverflow and reddit scrapper. And it still on chatgpts currently newest model, gets shit wrong or spins out promising the *real* solution is one prompt away to the point where anything beyond the most basic of functions I'm just telling it to explain the API or whatever to me rather than having it write any of the code.Right now, I can't plug all the information of the front end, back end, database, and integration flows into this thing. It's too big, it could not digest it, and how it's actually supposed to work is supported by people who know random bits of functionality and workflows spread out across the company - and this is for maintaining a relatively basic and small application mind you.I brought this up in an AI thread the other day, and some guy was like>well, AI can help task management and do the the documents for you!But that doesn't make any sense because how is the AI gonna know unless I just tell it myself? And at that point I would just write the documentation. All the AI is going to do is add a bunch of fluff to it and fuck up the formats.
>>533774282Unrelated to this reason.... I'm all for UBI, if it means killing off most other 'general' entitlement / welfare programs. WHile I'm a conservative, I understand that this is the only real answer, and untimately will cost less in fraud and administration costs.
>>533774282No they won't. UBI is never happening. 90% of the population will be culled so the rich can live with robot servants and a harem each. You are not part of the future, unless you do something about it today.
>>533775673I think we're beyond (in terms of capability), but seriously.... if you were an engineer for any AI company, would you release AGI? Knowing it'll instantly replace you? Or would you keep it in your pocket on a air-gapped mainframe?
>>533777625This 1992 book says it will be mostly chinksects when fourth international has unlimited power
>>533775362Corrrect
>>533778591AGI is impossible and even if it isn't, they wouldn't let it get to that point since it would be uncontrollable (aka, unprofitable)AI is fake and gay
>>533774282yes, a slightly better benchmaxxed model means the whole construction economy will move to building penthouses, lets start with mars, that was supposed to happen in 2018?
>>533778569So you are saying I need to get rich?
>>533779073agi is entirely plausable if we build it on myconoid hardware
>>533774282how the fuck its worse than 5.4 i dont get it, wtf
>>533779418What do mushroom men have to do with this?
>>533777565>for a token based service it will be profitablethis. literally the only way for them to ever be profitable
>>533777565>meanwhile deepseekv4 released, the moat gone and they're charging 20% of what anthropic tries togood fucking luck
>>533774282Ai is scam. Ai coding produces absolute garbage work. Posts like this are advertising from desperate morons
>>533775102sooonhttps://www.youtube.com/watch?v=rGfaJZAdfNE
>>533780322>good fucking luckIt's already figured out, like it's done with steel for example, for some projects, you will have to produce certificates proving you didn't use Chinese stuff
>>533777565>they willJust because, right? Anthropic doesn't even have a viable product and has 100s of billions in expenses planned out. And nobody has any clue that AI can ever do what they're promising it can do. It's a fucking guess in a new field with unteated tech and yet, with so much certainty, "it'll happen bro"
>>533780475soon....
>>533775863Quoting a jew isn’t making your argument better
>>533780486this isn't material, you can selfhost deepseek as it's open and it's just that much economically better it costs 5x less, retards at anthropic are pulling openai's trick from gpt2 era: it's too dangerous to release kek all fucking hypers and astroturfers, they literally have to viable way to profitability (they had before getting trump to ban their competition, muh dangerous model, muh china, now too late)
>>533780624>if it’s not profitable by next quarter it’s not worth the effort. >let’s make another iPhone insteadThis is the cancer killing the west btw
>>533775102Once they destroy the rest of the economy.
>>533774282>"real work"like more cartoon porn and blatantly wrong answers to simple questions?>>533775036>Have you seen what agentic AI can do?yes, cartoon porn and AI SLOP YOUTUBE VIDEOS SO YOU CANT HEAR THE PAJEETS VOICE>>533775863>first they came for the CS majorsso they came for the street shitters? thats the only ones in CS besides out of touch idiots
>>533780865Nah illiteracy and an inability to think is what's killing the west. I didn't say it's something that will not ever be useful, I didn't say investment shouldn't happen. I said that it's an emerging technology that is overpromised, and highly speculated. Anthropic is worth a trillion dollars right now based on pure hope. Trillions of dollars thrown at LLMs worldwide over the last 2 years by gamblers. It's a massive liability. Propped up by dimwits like you
>>533781237Well if it’s worth investing in and will be useful in the future then you can stop bitching and moaning about it all day can’t you?
>>533781490 Possibility of being useful doesn't mean that it will be. Investment with any emerging, unproven technology should happen at a much slower pace without weaving our entire economy into it. If AI fails to deliver in the next few years we could see a global economic crisis. It's foolish.
>>533774282claude does my work while I shitpost all day
>>533774282they say the same shit everytime. this faggot is a salesman
>>533781751Meanwhile countries that are more concerned about long term success than quarterly profits end up dominating. But hey don’t let your 401k drop that would be the real tragedy
>>533778305>No meme robot will ever replace me.That's a very bold claim in this day and age, wagie.
>>533774282What happens if you ask it how to make 100k in a month?
>>533781944they literally tried to and faggot robots either accept order of million of burgers at a drive through or offer 20k refunds, it's a fucking bust wherever they try, it has no liability, muh crack addicted Mary can be sued, good luck suing anthropic
>>533775036Yes, I've had to deal with the output of such systems. It's impressive in some ways, incredibly painful in others.
>>533774282This is just more of the same shit. Diminishing returns. Claude is better and will often seem impressively amazing until you realize how many mistakes it’s making and have to waste time addressing them. LLMs are a dead end. They won’t replace your job, but they will use them as an excuse in the meantime to eliminate jobs with the hope that they will. They won’t render you useless to the elite and culled although you better believe the elite very much desire that
>>533781237>>533781751it will be a crisis even if it does deliver. It's priced for absolute perfection, the companies are loaded with debt and not profitable and don't have a great path to profitability. Especially given the chinks are seeking to undercut them, deepseek 4 dropped today at 1/100th of the cost of openAI and anthropic. Build outs of useful tech that still caused bubble popping issues are the norm, happened with railroads, electrification, interstate highways and the original dotcom bubble. It will happen this time too. It needs to happen, to get rid of this wealth effect and unfuck computer parts markets.
These AI shill posts are pretty entertaining when you read them in Dr. Vindaloo's voice
>>533781035CS itself will probably be fine anyways. Well, besides the problem of H1Bs out the ass and offshoring which has always been an issue.A lot of companies who snapped back too hard after overhiring during covid aren't going to be in a pretty spot when all they have is a revolving door of senior and intermediate level engineers who don't stick with a company when a job hop results in an immediate 10%+ raise vs holding out at one company for an extra 1-3% a year. Some of the reaction was normal. It may take some companies realizing the limits of AI before a counter-reaction comes once they run out of SMEs.I am already seeing this phenomenon happen though. Companies are already getting desperate for seniors with no ecosystem of juniors to support it.
>>533774282Didn't he say that about 4, and everyone hated it, even vs 3?
>>533778916>>533775291I actually just learned it CAN count Rs now
>>533774282>...they will keep us alive, right?they are going to promise you a star trek future right up until they lock you in an AI dictatorship of no escape.
>>533776201>If humans can no longer meaningfully contribute to the GDP, what even is our purpose?We live for ourselves and those we hold dear, but I assume you mean what is our purpose for the elites.In that case, they need us to be poor and desperate. Nobody rakes in billions to eat caviar on a secluded island. That's much cheaper, so they do it purely for power over other people. There is no elite without the rabble.
>>533777565>just long enough to make sure local is dead and buried.So what's the plan? Make sure nobody does local storage/processing and then charge for cloud computing?Kinda like LA getting tricked into switching to buses for public transit and then the companies just refused to make spare parts so everyone had to switch to cars?
>>533785107fools errand, you can get gpt4.1 level answers on a desktop currently with the chink models and gpu from 2022, that used to be sold at fucking api prices 1.5 years ago and peoole were claiming coding is dead, apparently now you gotta pay few us companies or it's not dead, oh wait deepseekv4, anthro/openai are dead, anyone can selfhost it, n-n-no 1.5% on this benchmark means you gotta pay!
>>533775535What percentage of its text-based training data comes from reddit posts?
>>533774282MAID, vaccines, forced famine & forced war will take care of most of you.
>>533778389I am also a principal software engineer at a fortune 500. I feel like I am the only sane one at my company. All the management has completely bought into the AI stuff and all their direct reports are a bunch of ass eating yes men. Honestly AI is completely fucking useless. In the time it takes to create a markdown file to instruct the agent to do something, I can completely program it myself. Then after creating the markdown it generates a bunch of absolute garbage and I have to manually go in a correct it myself anyways. I can never trust the output code, it's always wrong and in the worst ways. It creates code that is not easily comprehended by a person too, so maintiance is a nightmare. It will completely refactor thousands of lines of existing code to enter a single assignment statement. The documentation it generates is not valuable to an actual engineer. It creates fluff pieces that satisfy processes created by people who don't do any work.The worst part though is it speeds up the workflow of the bad engineers. They don't review their code, they just create PRs and pray it works. Then I have to review these disgusting AI slop PRs and try to determine if it broke anything. If it breaks shit, I get blamed. It's never the tool being shit, it's always put on the people approving the final output.I fucking hate AI. It's not even slightly able to perform the tasks it's being sold as able to do. I hate the indians and retarded nigger managers who buy into it too. Its only use case is to act as a google search, and even then it's often wrong.
>>533774282No, there will be no ubi. They will kill most of us and those who remain will be slaves to be played with, tortured and hunted for sport.
>>533785850But bro, look at all the emojis in those mds! SICKSeriously though, at this point my immediately shuts down if I see emoji spam in markdown. I know it's not worth my time. As for the errors, as you've pointed out they cluster differently from that of a human making mistakes, and as a result EVERYTHING must be reviewed, and nothing can be trusted as one might in review of human output. Slop.
>>533774282depends on what you mean by "alive" and by "keep" and by "they"
>>533774282hehe
>>533785850Im mid level at this point so youd know better than me but yeah, I struggled just trying to get it to write me a throwaway dataweave script until I just gave up and had it regurgitate the mule documentation on it in a slightly more digestible fashion.I imagine most of it is pressure from above to use it just to appear on the cutting edge or whatever and damn the actual functionality. It was the most hilarious thing at a company wide meeting where they were trying to pitch us on how we've been using AI at my new company and it turns out what a bunch of engineers got pulled away to work on and come up with was just an internal HR chatbot to save time on employees asking about PTO protocol and such. Which will probably inevitably end up useless when our HR gives up on plugging instructions into the thing. At my last company it was worse, I had to try and use azure AI to generate test steps and integrate it with selenium shit. Basically read one human analyst excel file and convert it into an excel file selenium could read. And it just could not be done. The output never was consistent with delimiters, it could not actually scrape the web to make up steps itself, it would follow all my instructions with examples far too literally, just an absolute shitfest.
>>533782113>nooo! it can't draw hands yethere here
>>533774798If you can't into hex editing and hook injecting, you don't deserve to cheat on vidya, subhuman.
KILL YOURSELF
WHAT DO PEOPLE ACTUALLY DO WITH THIS AI SHITWHAT IS ITS ACTUAL UTILITY
>>533774282the logo looks like hands rubbing
>>533774282>they will keep us alive, right?Yes, to power the data centers
>>533787905literally nothing, 95% of attempts at introducing LLMs into corporate workflows end in failure, it's all just: but in two weeks it will xyz
Anti AI people are so dumb. If you had any actual convictions or beliefs, you'd go full Uncle Ted. But no you want all the other technologies that caused massive disruptions and widespread issues, but this one is where you draw the line? Actual woman brain.
My company already replaced a ton of people with AI. Most recent was all the accounting and expenses guys, people who managed the expense reports, audited receipts, that kind of thing. AI does it faster, cheaper, is always available, and makes less mistakes.
>>533788553yes sure, and then IRS asks for actual.data not hallucinnated one and your imaginary company just says: haha, it's algorithms bitch can't touch this
>>533788722They still have some people in roles that are basically just supervising the AI and double checking important stuff. It's not entirely automated, but mostly. I'm not too worried about it, wouldn't mind if AI automated my computer tasks, I mostly do fieldwork installing/repairing machines and AI won't be doing that for a while. And most people are too dumb or too weak/fat to do what I do
>>533788553AI hypers pretending to use AI in finance is the funniest, it's the thing that will rape you the quickest, no amount of: but ai said so will save you from IRS
>>533787341We often get massive excel spreadsheets of conditions in tables for determining customer risk at my job. The formats are always inconsistent and we usually have to convert all this data into json documents or sql tables. I tried using AI for this task once and it was injecting fake data all over the place. I spent hours trying to get this shit to work and had no confident it was right. It took my 15 minutes to write a small script to just convert the files.
>>533788863so you keep your head count, bravo, you just increased the spend with 0 savings, literally the same story with AI everywhere, ppl having to manually recheck for hallucinations, extra spend, 0 saved, it will revolutionize economy my ass
>>533775102Never. They are massive jew equity scams.
>>533789087You have no idea what you're talking about, man.
>>533775596Yeah, well, it has always been the case, it will still be for a long time.But to a midwit, an LLM certainly looks intelligent, even though it is just statistics over the words we use.
>>533775036You don't know what you're talking about
>>533774532how big of a boat do I need to reliably sail me and five buddies with supplies from baja to new zealand?
>>533775863"AI" don't improve, they slop. And the human using AI slop along. It's a long fall to madness. If you fall too much, you suicide.
>>533781490>JUST INVEST IN ANOTHER DOTCOM BUBBLE OF BULLSHIT BRO!!! YOU'LL BE WORTH BILLIONS BRO!! JUST SIGN EVERYTHING AWAY!!!
>>533774282Imagine paying openai to know all your company information. Incredibly stupid.
>>533777565>just make sure no one else has it, but we have itah, the NFT economic model then
