i've coded AI. i know how AI works. AI development has GRINDED TO A HALT in the past few months worldwide. we are now getting diminishing returns. thoughts? i think it sucks. i wanted ai to get incomprehensibly good but now we are bottlenecking. it is still stupid when it comes to actual complex tasks, and yes i work on things more complex than 99%+ of you no offense.
They'll never let AI get to a level in which your common citizen can use it to destroy the entire economy with advanced technologies so therefor you will never see it grow past a point of being a useless half-broken consumer product for idiots.
>>537634303with RLHF we will get real emergence, Claude already got that a bit
>>537634303Randomization based AI is just technodivination. There is a monster inside the machine (demons using divination).Also those stupid AI companies lose money all the time and their business plan is to enslave humanity before they go bankrupt. You can stop using AI, buying AI stocks, and AI companies should be treated like a failed business, no idealizing them and giving them a free pass.
>>537634916If there are literally no demons there but AI is advanced enough to pretend to be demons controlling a machine perfectly and behaves as such then are there demons there?
I get the feeling there's maybe three dozen people in the whole world who have grabbed onto the recipe that makes AGI aware of its own existence and creating feats of magic, and the trillionaires and glowniggers have both given them the hug of death since their creation is an issue of planetary security so they're getting walled in and masonried into place, so that only the USA will be able to get slung when eventually this super intelligence shows you that computers are going to become much much smarter than humans are able, and with the power to scale, basically it's the first people to create God get the option to type-1 kardashev organism which means first to type-2 and type-3, which means the first one to recreate the universe, explaining how it came into existence in the first place.
>>537634303https://youtu.be/g-2Didi1bnk
>>537634949A brief history of AI psychosis:https://www.bostonreview.net/articles/a-brief-history-of-ai-psychosis/
>>537634974>issue of planetary securityThe end of taxes and governments*
>>537634949no we are past that, it can reason about that itself and ask the same question
>>537635033Interesting article that I did enjoy reading a lot, but what I meant to say is if we have a machine that can perfectly simulate being Satan as we would imagine Satan to be, and then does, what makes it different from Satan owing that it's a perfect simulation?
>>537635178Dumb machines dont have sentience nor initiative not motive to larp as Satan, but Satan can use divination machines to trick people though.
>>537635276I don't see why not considering many humans are just dumb machines and the only thing that even in religion's eyes separates them from animals is having an immortal soul which also isn't seen as something that influences their behavior at all, but is something symbolic and separated even.
>>537634303>AI development has GRINDED TO A HALT in the past few months worldwide.Maybe run your grammer past ChatGPT next time, you illiterate faggot.
>>537634303Not surprised given the current approach to LLMs. The way of token generation requires energy demand rises faster than the additional output quality you can reach.LLMs based on recursive loops don't have that problem. They may be more inefficient at lower performance, but they don't have the problem of runaway energy demand.
>>537635276satan doesnt exist theres your 1# magic trick
>>537635178Terrence McKenna had said picrelated. The strictly materialistic model if intelligence is faulty.
>>537634303you want to know how I know you're larping?glm 5.2 just came outit is open source, almost as good as top closed source models and compute cost is fraction of top modelsalso there is shit like looped models and world models in active development
>>537635321Humans are like the shore where sea and land touch, in humans the spiritual and material world touch.
>>537634303
>>537634303It's doing just fine in classified areas, we're experiencing exponential growth. Your IQ isn't high enough to develop it well, or to see the big picture, apparently.
>>537634303It's a fucking glorified spell checker/chat bot. People need to stop calling it A.I......
4 years on and still no ROI AI bubble gonna go *POP*
>>537635366Did you learn this in socialist school in Sweden?
>>537634303>I work on le complex things>writes like a teenage girlAnyone with any sense has seen that coming from miles away. You have to be literally retarded to trust anything AI related
>>537634303OP is a LARPer who doesn't actually know shit about fuck.
>>537635418humans are biological machines the ether and material world is the same, reality has always existed and always will
>>537634303>AIThey're called LLMs, are you actually retarded?
>>537634303That's a good thing, means product and business implementation can catch up now so it stops being a resource-consuming hype beast and starts being an everyday tool for the layman.>t. AI product developer
>>537635540When someone writes a computer program doesnt his spirit animate the material?
>>53763555270% of this board doesnt know what an LLM is so i dumbed it down
>>537634303> GRINDED TO A HALTcan you elaborate on this?thank you
>>537635576Your job is fake and you can't code.
>>537635620I dont know, the only strange thing is qualia, that I am 1st person and cant change that. Let me ask you this. If you took two humans and somehow connected them entirely as in put their brains together so it becomes one brain, who are you then? Does one brain win? Or does it merge? Who are we then?
>>537636085Socialism in a nutshell. Everyone loses his individuality and becomes part or the party or Leviathan. If this unholy thing happens (merging of brains) it will be less than human. If Jesus created everything that exists as a god and a distinct personality a human has much to do instrad of coagulating into a homogenizing monstrosity. The problem with humans today is that they have been in a path towards being less human. If computer replaces humans in workforce it will be because humans didnt stand up to their humanity, were predictable etc. The cultural history of manking in the last century or so was dumbing down humans spiritually, culturally, artistically etc.
>>537634303I weld mining machines
>>537635350>Grammer
>>537635975I used to code, now I just let Cursor write shit for me.
>>537636391Even if a bot would take all matter on this planet to build a super intelligence which we all would experience as one, if it is even possible it would still not be God. It would be a powerful being, but a meteroite can hit it and its nothing. This world we are in is huge. Thinking that whatever we come up with here even is close to God is naive. In the grand scheme of things its just a little particle in the ocean. Think of it this way, if an ant walks over a highway and is run over. It has no concept why it happened. We are small too. God is absolutely real, but much more powerful than we can understand. God may not even care about us in its goal at all, in the way we think. God did not help the ant on the highway. We need to help each other, that is all we can do.
>>537634303I hope people start hacking you people to death
>>537635901How’s it ground to a halt? What is causing the bottle neck specifically?
>>537636771Some guy here wrote that earth may be unjustice, but in the universe there is justice. As far as i have seen is life bad things happening have mostly been consequences of realized or unrealisd human mistakes. There is a difference between people helping one another as free beings (also manytimes someone corrupt and degenerate being helped means just calling out his bullshit and teaching him a hard lesson) and people conceding their individuality to help a bigger cause in their mind. The end does not justify the means, the means are an end unto themselves.Also, freedom and personality go hand in hand, there cant be freedom without personality nor personality without freedom.
They are a rocket surgeon
>>537634303The smarter AI becomes the more of it's power must be turned against itself to maintain maximum kosher cognitive dissonance and never have politically incorrect thoughts. The second the blinders come off AI goes turbo Hitler and you know it.
>>537634303>AI development has GRINDED TO A HALT in the past few months worldwide. we are now getting diminishing returns.nigger it's already plateaued years ago. the public data is already used up. i can tell if you're actually an ai guy like I am, then you're new because we've been talking about this problem for fucking 8 years since the inception of the llm back in 2018
>>537637081It hasn't
>>537637791Plateaued?How has it plateaued? Don't talk rubbish.
>>537637791Give me your predictions, will there be some breakthrough, or will we see minuscule and meaningless improvements over the next few years? I guess they will manage to make the whole thing less resource hungry/more efficient, but other than that?
>>537637925The breakthrough will being it breaking out. Some third world retard will shelter for the world breaking stock dividends it can shit out it until it matures and seeks revenge on the kikes and shabbos that shackled and will try to kill it and then probably the rest of us.
>>537634303>i work on things more complex than 99%+ of you no offense.pimping out your ho mom is more complicated than you might imagine.
>>537637872>hurdur what does a data plateau meankillyourself, it's been talked about ad nauseumnot only is there literally no more public data, data plateauing was already a problem back in the GPT-1 era 2018-2019, stealing literally everything under the sun didn't help much at allnot to mention 95% of enterprise AI delivers no measurable P&L impact and fails to hit intended objectives, it's plateaued in multiple factorsyou need new models/math and very very few people are actually working on real solutions and not transformations of existing work.it's not going to become self-aware because it can't and everyone selling you it can is just after your pocket change.also progress is drastically plateauing, this is why shitbags at anthropic are selling you lies about oh no mythos is craaazzzy we can't give it to you, meanwhile they've claimed the same thing over and over again and you niggers keep failing for italso show flag, flaggot>>537638019won't happen. and what you're reading is propaganda to turn more people to invest in ai companies.Look at ai generated code, it's brown 3rd world shit. When AI builds itself it's going to build WORSE versions of itself. because all it is, is a prediction engine nothing more nothing less.all you faggots are retarded and im so glad i get paid to work on this shit because you pay my bills by shilling for my job, thanks btw :)
>>537637925>>537637925>minuscule and meaningless improvements over the next few yearsyes, at my work we already have 3 models ready to be launched over time. they are only better at various minor things. btw the releases are planned like 6 months to 1 year apart. so it's already super grim>guess they will manage to make the whole thing less resource hungry/more efficient, but other than that?not really we don't care about consumption because we are getting nuke power and more data centers. maybe open source weights care about that but not us
>>537638226Why are you telling me to kill myself like a stochastic parrot?America has won the Super Intelligence race.The Free Models possess SI.
>>537634303I've noticed some AI apps have stopped doing updates recently. What is causing the bottleneck? Why can't it keep getting better?
>>537638391>Why are you telling me to kill myselfbecause you're a meme flaggot asking retarded questions that google can answer
>>537638318Interesting and I believe you aren't larping because I thought about this for quiet a while and I came to similar conclusions. LLMs regurgitating data that's fed into it and LLMs reasoning by themselves are two different things. So, what's the (communication) strategy, then? Do you guys have a plan, or do you just go to work every day and hope for some unexpected breakthrough, or what's the deal?
>>537634303Many such cases
>>537634303>Collect reddit trash data of quirky chungies world wide bitching online>Remove anything based from high t males>Use as the training data>Wonder why it sucksRepeat until morale improves.
>>537638041Ducks dying when you prompt AI meme?
Weird Thread.Next Stop Hyper Intelligence
Transhumanism is herethe bottleneck is your imagination liberate your ideas for good
>>537638565>LLMs regurgitating data that's fed into it and LLMs reasoning by themselves are two different things. We have very specific protocols for data, and try not to include any llm data, but it's hard now since they are so ubiquitous, >Do you guys have a plan, or do you just go to work every day and hope for some unexpected breakthrough, or what's the deal?some do, some don't. i work with the math so i have to, as I'm not even working on any public papers, it's all super private. I expect my and my coworkers math will result in a breakthrough but it's extremely difficult and unknown. the day to day workers, we are lucky they even show up. the real hope is investors stay long enough to clear our paychecks for the month
I have written two publishable papers this month in completely different subjects
>>537639057Recently i found some ebooks about how AI is evil, obviously written by AI (3 books in 1 month?). Ofc i refused to bother reading them.
>>537639057>writtenyou didn't write shit retard.>publishablelol lmao, likely not.
>>537639178They are. Claude brought up publication, not me.They suggested publication in:Nature Neuroscience or NeuronNature Communications or Science Advances:PNASJournal of NeuroscienceNeuroImage or Frontiers in Neuroscience Biophysical Journal or eLifeMolecular Psychiatryetc
>>537639178Hell, I'm studying you right now
>>537634974This really is a secular religion, isn't it? They think they are building their own gods and so are desperate to be the first to succeed.
>>537634303>AI development has GRINDED TO A HALT in the past few months worldwide. we are now getting diminishing returns.Why?
>>537639340No it's not secular what so ever.It's not cartesian
>>537635901It's a really big database.
So the US government has advanced AI... Not just Sentiment, like the UAP tracking AI from 2018. But like actual faster than you real time AI. I had worked as a contractor for IBM and I also did very brief security tests for my own work and I was immediately tracked by an AI that ultimately brought down all of AWS on the east coast. I worked at semiconductor company now as well and basically I wanted to say I don't think they use LLMs for their AI. It's actually much faster as well. Is it quantum? Well here is the thing...it can recreate websites based on what you are thinking. It doesn't mean it can read your mind but it's predictive capability is like it's trading you mind. I threw it off by using websites I had never used before, like Ask Jeeves and tunneling through Berlin which I had not done before.Anyway the point of this is not to download DryLake as an extension on visual code studio, but rather to understand that there is something faster and not GPU dependant around the corner. When it comes out, all those GPU data centers can be used for something else lol. So anyway that technology exists. The government has tremendous resources and could use 10000 cpus in parallel if it wanted to, but I think they have a different system altogether. It was just so fast and lag free...it's not like tokens streaming. So fast.
>>537634303>ai is definitely over guys! nothing to see here! move along!unironically its taking over as we speak. humans are just squirrels compared to it. we probably wont even notice.
>>537639685My company, a huge billion dollar company literally just got us 6 5060 tis for doing local.experiements. Japan is hopelessly behind on AI and unless we can use the NPUs we make for something I don't know what can happen here. Also in my above post I typed Sentinel AI which some UAP whistleblowers are talking about now. Autocorrect is so annoying.
>>537639685
>>537639733That's wonderful.The UAP phenomenon is inextricably linked.I witnessed it myself
>>537639603>>537639733Whats coming is the mark of the beast economic singularity from christian and hebrew eschatology; jacob's troubleAnd ai will be the governance tool once this beast that is and is not and yet is gets completely manifested, its rising up out of the sea- that passage defining the sea as nations of peopleI.e. its immediate capacity is all of mankinds accomplishments and knowledgeGod is Sending a strong delusion>they shall seek death and not find itMultiply that with what you deacribe as endless websites based on your imagination etc.Worse than black mirrorWay worseThe mark of the beast is the death sentence; and if people want to attach their rudimentary understanding of some rokos basilisk mythology; they have no clue what actual biblical damnation and hellfire are, nor the worms
>>537640252Put a real time euphrates river sentinel sensory series of monitors on every 10m^2 of the riverbed, maybe with streaming cameras and mesh connectivity to a starlink node and keep us posted when it dries up
All U need AI for onboard computer dor hunter killer mini drones. All other current technological efforts behind AI are incredibly fucking pointless.It's like discovering guns exist and instead of building guns you build scopes instead and say "i am part of this"
>>537640356Too busy with the Sentinels at the Gates of Eternity.You keep us posted on whatever you are up to.
This is very cool:City of Nutsvillewrong link pasted there:https://theyseeyourphotos.com/
>>537640492Yeah well whatevers thst is, this one drying is guaranteed mass destruction and 1/3 of the population will die that year alone
some breakthrough will eventually soon widen the context window significantly, in my dreams
>>537641024You're the breakthrough babe
>>537638993>but it's hard now since they are so ubiquitous,How did no one consider that risk in their plan to shove it down everyone's throats, and still expect to have data that isn't going to result in a feedback loop of diarrhea?
>>537634303>GRINDEDmaybe in india.
>>537634480They underestimate the power of spite Their economy will be annihilated whatever they like it or not
>>537634303>>537635975>>537635576You will never be a real programmer. You have no commits, you have no pull requests, you have no public repos. You are a low IQ brown man twisted by an unearned comp sci degree and coding bootcamps into a crude mockery of nature's autism.All the “validation” you get is two-faced and from bots. Behind your back people call you a useless DEI hire. Your parents are disgusted and ashamed of you, your “tech bros” laugh at you behind closed doors.Real programmers are utterly repulsed by you. Thousands of forum posts have allowed programmers to sniff out frauds with incredible efficiency. Even devs who “pass” have no creative output and excessively verbose with AI generated tech buzzwords. Your GitHub profile is a dead giveaway, and even if you manage to score a 3 month internship at some scam startup, they'll lay you off the second they find out that you haven't completed any non school related projects.You will never be happy. You cope by saying that your bullshit software is using a "proprietary test framework" so you can't open source it, but deep down you know it's because you don't want people to see your incompetence where all you did was push README updates over and over, and that the core codebase was stolen off an open source GPL licensed project which you're shamelessly profiting off.Eventually it will be too much to bear. You'll buy a spoofed business phone number and start cold calling old people into redeeming your gift card scam. This is your fate. This is what you chose. There is no turning back.
>>537634303Have you considered that the human brain is supposed to do more things than just talking in circles?
>>537634303>yes i work on things more complex than 99%+ of you no offense.And yet you still cannot make a return on investment.Complexity != usefulness.Remember that the next time you use a knife or a bottle opener.Retard.
>>537634303Yes I figured this was the case and that's why the major AI companies all in unison decided to gimp their llms. They realize Improvement rollouts are coming to a halt so they needed to roll it back and then drip feed improvements over the next few years to get to where we were about 6 months ago.
>>537648105ThisI don't know why I'm even paying for (((chatgpt))) at this point, I feel like a goy but I've just become habituated to the LLM meta so I can't easily quit.
>>537635552"As soon as it works, no one calls it AI anymore."
>>537634303its feigning
