I am programmer but I just can't stand anymore the cringe of many software developers coping with the AI affirming without any proof that "AI won't outperform human senior programmers in the next 20 years" or some bullshit like that when the evidence says that it's gonna be only 2 years in the best case. (Maybe we are already there but the AI companies haven't release that info.)
So i have 3 theories:
1. There are "seniors" too dumb to understand what "exponential growing" means, and that is the reason cuz their stupid arguments always revolve around pointing out the current flaws of GPT4 what is completely irrelevant because the only thing that matters is how GPT 5, 6, 7, 8, 9, etc, will evolve. Nobody cares about your error analisys of GPT4, cuz that change nothing.
2. There are "seniors" too nacissistics to accept that now they will be more disposable than a plumber or taxi driver. If that is your case go to touch some grass
3. They are courses sellers. If that is your case your opinion is automatically dismissed.
> This topic is political.
most humans are morons
what's so hard to understand about that?
if you have to ask, then you're definitely not particularly smart yourself
it's pretty simple actually.
1. LLMs are "exponential" in the wrong direction for solving code; they can do basic examples because they've ingested working code and replicate it. but there's no internal logic actually writing code. the exponential growth is in replicating human speech, not in designing a codebase.
2. every GPT artist seems to be a beginner, where instead of asking an AI to write a function, an expert can just write it themselves. the best use case is helping experts save time, but even this is a stretch because...
3. ...GPT is locked behind a paywall. coding is fueled by the passion of FOSS, writing your own code, making your own work happen, and interacting with an API to abstract code will never be something that people are passionate about. it goes back to #2, that the adoption of GPT is only really led by the people lacking the skillset to do the work themselves.
there are other reasons that the exponential nature of GPT is irrelevant too, but i've already written too much.
Coping. You should read about the latest papers, the AutoGPT, the hugging, etc.
Is not only OpenAI, Google, Alpaca or whatever. It's about the thousands of open source projects popping up to tailor GPT to turn that into a baby AGI.
>tailor GPT to turn that into a baby AGI
now that you've completely abandoned your original argument, you expect me to address this too?
Adress it, coper.
Literally the Aligment is the only real limitation to unleash that shit. If dev job remains is gonna be only cuz of that, but yet AI will still outperform seniors for sure.
But if you don't believe watch this video:
alignment of AI output to decisionmaking isn't new. you should be equally impressed with speech to text. think about all the things that take voice commands nowadays. GPT is trendy and i reflexively dismiss anyone claiming GPT is the dawn of a new age. did you forget about algorithmic content selection? like products on Amazon or videos on Youtube? where were you when those incredible innovations were made? why is GPT the end-all-be-all for you?
Because those things were not even close to AGI, you massive retard
go ahead and define AGI for me. i'll go make some popcorn.
AGI refers to AI that is able to learn, on its own, to solve any intellectual problem humans are able to solve
not very hard to define at all
do you like acting like a retard?
My opinion is irrelevant, the only thing that matters is how i am already seeing that change in real time. How this cut more an more task of human cognitive work every day even in my team.
And you don't need to be fortune teller to know that. I just saw so many "AI Gurus" who only 6 months ago were saying that "AGI will take at least 50 years" with their head nailed in their ass of shame cuz nobody trust them anymore
>My opinion is irrelevant
ok, i am not reading the rest of your post, fuck you, fuck you for making this thread, go be a stupid moron somewhere else
> i am not reading the rest of your post, fuck you, fuck you for making this thread, go be a stupid
- Ad homimem
- No arguments
- "I dont want to read more cuz my avgina was hurt"
- run from the debate like a pussy
Screen shot to remeber this victory againts the gayest Anon ever.
You're a autistic retarded gay. That guy is above you in every way
Ok. He is cuz you said.
Trust me bro
Your arguments are sound but your writing sounds like an idiotic moron trying to get some approval from other people in a website with no voting system, is this social media brainrot or does no one love you (including your parents) irl?
Either way, go make a post in reddit if you want a "circlejerk"
Eat shit you Steve Harriston Quispe Mamani.
You should by happy cuz for first time you will be able to fuck AI women who don't look like pic related and you will be able to quit gayry.
fuck off to r*ddit dumb npc fucks
Damn you're really upset aren't you little retard.
imagine getting bodied on 4chan by a midget third world spic
I used it to populate a database with some realistic mock data recently. Pretty good
I made up a pretty complicated representation of data using some inconsistent formatting (stuff like <number1>x<number2>_<label>|<number3>, ...,), then described the general sort of data it represented and asked it to tell me what each value represented. It correctly inferred what I was going for with every field. It also generated several solid alternative data models for the same information when asked.
It has no problem writing code that hasn't been written before. You are coping and will be out of your overpaid gay job within 2 years.
sure...ask it to write a shader that runs dynamically if/when you have a set average of light in your vector space to optimize performance
I tried autogpt and the likes, it's useless in writing serious non-cookie cutter boilerplate code
ask it to write a root finding function that uses newton's method without just importing numpy
There are actually examples of it online, but chatgpt incorrectly copies the information and changes the variable names to try and seem more convincing LOL
>find the root
You realize root means whenever a function equals zero, not the square root of a number
Also why can't it take the square root of a negative number? It could just take the absolute value of the number, find the square root, and then slap an i on the end of it
Thanks for the laugh bro
You're not a coder. Programming languages hardly ever support complex numbers natively.
If you are going to shill for ai, you should probably learn some math, and also test the code instead of posting the bullshit it outputs.
You're assuming that because it generated some text, that it's error free and ready to be implemented
good night robot shill brah
stop being gay
Keep the good job Anon proving to these gay deniers that they will need to learn welding eventually to avoid end up living under a bridge
>LLM are different from actually writing code
speech and code is for all intents and purposes the same - its just structured data transformed in steps - with enough processing you can break down every process to (logical) steps - be it love, emotions or just some fucking code
>speech and code is for all intents and purposes the same
Do you know what writing code requires?
KNOWING WHERE YOU'RE FUCKING GOING.
When you look into how chatgpt REALLY generates text, I'm sure you'll understand why getting it to 'write code' for solving real problems is not gonna do what it says on the ad.
>GPT is locked behind a paywall. coding is fueled by the passion of FOS
> He doesn't know
you're literally just parroting the exact moronic nonsense OP is calling out
Some people just can't see the future.
We haven't even been able to build a 100% self driving car you dumb fuck, there still are problems like rain and snow. And you think we will next GPT will be writing complex code?
GPT is just stackoverflow on steroids. It will increase productivity for sure. Whether that will lead to mass unemployment or not remains to be seen. Based on what happened after things like compiler optimizations, IDEs and stackoverflow, I will bet it will actually create more opportunities than take them away.
Do you understand "Exponential grow" gay?
Anything that happened before of March this year is completely irrelevant idiot.
So that, shit now have capability of self-researching, self-reflecting, self-debugging, etc, but won't be able to build and stupid backend and fronted from scratch? stupid micrservices?, a stupic MVC?, stupid videogames? Daajj!!
Stop your gay cringe .
you stupid morons keep coming up with fucking buzzwords and shoving them in our face
"It's political nya nya nya"
You're so fucking annoying
"Exponential grow" sounds like some sci-fi shit
take it to BOT and tell them you want green light
are you two having a competition to see who can out-stupid the other? Genuinely curious, or is this some kind of advanced roleplay larp?
>Do you understand "Exponential grow" gay?
It is in no way exponential when the limiting factor is the human perception necessary to discern whether you have meaningful content generated by the "AI", categorically wrong.
AI is trained in dialogue with human beings and prompted only at behest of human beings. It cannot do anything by itself and it cannot do anything original. It will always be one step behind no matter how quickly it generates text.
And besides, the most damning thing of all is this: it's trained on reddit content and moderated by Redditors. That makes it incapable of doing the one thing it was supposed to do - truthful summaries of factual information.
>. It cannot do anything by itself and it cannot do anything original
hmmm, yeah you lost me here. Baby AGI, AutoGPT and the Microsoft thing, although incredibly lame and early days are clearly heading in that direction with chain of thought and inner dialogue work done in the background. A lot of people are taking these things quite seriously. I am hesitant to say it isn't possible with current tech, and maybe it is already working somewhere
You are definitely the uneducated one here.
Surprise surprise the people writing papers about AI who are trying to get funding for their AI research super pinky promise these investors it will replace their workforce and save trillions.
I'm not saying if it's ethical or not, i'm only point out the current capabilities and future capabilities. And the fact that it will meake the top 1% even more rich, is the reason cuz i absolutely think the growing will be even more exponential.
What could motivate the invest and acceleration of AI more than fuck off the worker class and make the rich richer?
I do hardware/firmware. Unless the AI can build PCBs, i'm good for the moment.
>BAZINGA THE COMPUTER ADVANCEMENT IS HYPERBOLIC WE ARE GOING TO BE SO SUPER ADVANCED ITS GONNA BE INSAAAAAAAAAANE
>all of the infrastructure is 100 years outdates and crumbling
>roads cant even get fixed, major interstates have potholes
AI can be as advanced as it wants it wont change the reality that we already live in the ruins of a dead civilization
no (you) as always
eggheaded computer technology moroncattle golems stay seething forever worshipping daemon ex machina meanwhile everything is already dead and collapsing around you in real time in the REAL MOTHERFUCKIN WORLD
wake me up when your technology does something that changes the trajectory of this dead civilization (it wont)
>just wait we already have hypertech in the dumbs, 2 weeks until the new world is born
keep me posted
>I’m a programmer
What’s your backup plan spic? We’re about to legalize cocaine soon.
> We’re about to legalize cocaine soon.
Those are great news for our GDP. I hope other chink countries follow your lead.
> great news for our GDP
Yes but you better put down your keyboard and start picking cocoa leaves.
That is my secret Chang, i never stop of picking cocoa leaves
Kek, you act like chinks haven’t invaded Colombia too, I’ve been to several chink restaurants in Bogotá.
I'm not an expert in chink food
The soup of dog causes me diarrhea so i avoid those places.
you can't be serious
cocoa is chocolate
coca is cocaine
you're actually letting people derail your thread
have a bump
4. They are in a death cult and want to create their god through AI and annihilate//enslave humanity.
>The son of man coming on a cloud
It only replaces midwits and lower.
Geniuses will continue to do fine and be highly sought after.
It is about time that people realize qualification just prove you are capable at the minimum level but they don't actually show how intelligent you are or your intelligence outside your specialist subject.
Ai only replaces those who can only regurgitate information.
Hubris is dangerous... What will any human be superior at over AI in 100 years from now? You also should realize that the elite want to fuse humans and AI together.
>You also should realize that the elite want to fuse humans and AI together.
This is a more plausible scenario, GPT will not create an general purpose programmer AI, but through could BCI could seriously boost productivity of good programmers in the market.
there will not be a wagie apocalypse, if such illiterate thing was possible then the capitalists would have heavily invested in AI hardware in the 80s, instead there was absolutely no funding despite back then AI technologies showed the same promises as of today, but still we had the AI winter and much research and investment was halted.
I can't run as fast as a car but that doesn't mean my legs are useless.
Just because Ai can do tasks it doesn't mean humans are totally replaced. Who is going to provide the innovation? A machine that requires data to learn first?
I'm not watching a 15 minute video just to be able to reply.
I have heard the arguments before and Ai taking over jobs before it is ready is a real concern as is having all the humans too stupid to give it anything beyond a basic prompt becoming unemployed due to their lack of capacity to learn.
It is a problem for midwits because the exceptional will adapt and its not like the machinery of old which was inaccessible to the poor person it is readily available so only those without the brain capacity will be unable to use it.
Say Ai is 100% capable of programming does that mean it will spit out trillions of good quality products all on its own without any human intervention? It will just copy things it already knows.
Even if it replaces all current jobs it doesn't mean humans are worthless it just means industries need to change just like introducing early production line machines made industries change.
Ai death machines hunting down all of humanity are a real threat but Ai replacing jobs? It's only a concern if you are midwit or lower.
> I'm not watching a 15 minute video
>be able to reply
Go back to Tik Tok.
No comments because you are midwit or lower.
You require me to watch a 15 minute video because you can't even put it into words.
You require someone else to explain for you.
You are exactly the kind of person I was talking about, You might have skills and knowledge but you cannot adapt nor can you provide value outside your specialist subject.
> I'm not able to watch 15 minutes with the latest papers and updates of Open AI cuz im a fucking lazy ass and retarded.
>Lack of information enough to reply
>Unlike me you who bother yourself in watch the 15 minutes of video are the mediocre one "midwit", while i am super smart.
> I assume those conclussios are yours instead of Open AI cuz i was too retarded to watch a short video of 15 minutes
> I am yet smarter that you. Trust me bro
> did i say tha you are "midwit" yet?
A lot of people are incapable of even attempting to use foresight. They are in the here and now and nothing else matters to them. Cognitive dissonance triggering plays a huge role in this.
This is disgusting
trying to shame people into watching worthless propaganda from sociopathic Silicon Valley freaks like Mark Zuckerberg
have a nice day, gay
have a nice day
>You require me to watch a 15 minute video because you can't even put it into words.
Anyone who call on halt on the research of AI is an outright retard or illiterate, and this kind of things makes me quite skeptical about academia members signing shaddy documents, yesterday I saw an article in the press, why we must halt AI research because:
>An AI algorithm succesfully detected faces of white male and female faces but failed 70% of times when trying to detect the faces of Africans.
Really? PoC is the great concern of these academic morons?
This AGI thing is non-sense, with the current technologies you dont have the hardware nor the software required to do the computations, Artificial neurons are "Modeled after" not "imitate" biological neurons, the computing power does not exists, because hardware has not been designed to do such computation at a cost-effective way.
And what the fuck is this "AI Alignment researchers" is this another bullshit job like "Senior Marketing research and costumer behaviour analyst (they/them)".
People are throwing pseudo-scientific and pseudo-technology jargon and pretend to create a smart system.
I watched the 15 minute video. Watch most of his stuff and he doesn't fuck around. My take is it is all very interesting these hot takes from all over the industry, but the fact remains, the horse has already left the stable and there is no putting it back. All these things about slowing down and maybe it will slow China down seems incredibly flawed logic. It is entirely feasible China is ahead. Makes you wonder when 90% of the OpenAI team are chinks ffs. Stopping the GPU's will probably happen, but not to stop nation states, more like just to be asshats. I am pretty sure training large language models at scale doesn't even hinge on GPU's. It's exactly what I'm working on, and even though I haven't done it yet, I bet someone has or will. Anyway, I'm a senior, and have been using it for a while as a productivity boost. There are definitely going to be lots of people laid off, and me included. I'm already trying to adjust my thinking to doing stuff more for fun than anything else, even though putting high level people out on the street probably won't happen (due to logistics reasons), for a while
>It is entirely feasible China is ahead.
The Chinese are way ahead with their 1978 model using the latest GPUs that were bitcoin mining until a few months ago.
They're so far ahead they're almost 40yrs behind us.
>The Chinese are way ahead with their 1978 model using the latest GPUs that were bitcoin mining until a few months ago.
that was my thinking
>They're so far ahead they're almost 40yrs behind us.
not entirely clear what you are saying here. I think you are saying they are doing just fine for an ancient model, if I read that correctly. There is something for that if it is running their country already I suppose,but good to hear the lads in the back room are cooking up a treat
usa < China < British
>usa are lacking ideas
>china are lacking ideas stolen from the usa but have access to SotA hardware
>British have world-beating ideas that work and represent the future that work on even old cardboard PCs from dumpsters
Welcome to 2023.
>keeping it SIMPLE, unlike daft americans who would moron it to death with complexity.
kek, right. very true, there is something good to be said for always having resource constraints
>I did the math for the model and the VERILOG for the digital hardware representing that model.
that is awesome anon. I never did get into VERILOG. sounds like a pretty cool thing you have going on.
>It is entirely feasible China is ahead.
Such wasted digits... No... china is at best 20 years behind the west, their biren chip is good but its not an AI accelerator it is another GPU.
The best they can do is copy opensource projects at this point.
You're spamming worthless yt garbage
have a nice day
we hate worthless yt spammers like you
Then go to read the latest researching papers directly in PDF you retarded moron.
i kindly found a video with that info translated for normies cuz i doubt many idiots here could even read, but you don't appreciate that so go to brush the hairs of your ass inwards then, you massive gay.
that's just spam
research paper spam
actually they put them on "arxiv"
you're spamming yt assholes who want you to buy into their yt spam drama
spamming yt isn't drama
it isn't news
you're full of shit
you're a goddamn spammer
Innovation for what purpose? The AI will probably have some goals that humans can't even understand or offer innovation towards. If it isn't sentient actually then it will just be a replacement for cattle work labor. You realize that cashiers 50 years ago never imagine being replaced right? They targeted people like cashiers because no one gives a fuck about them. This is how they manufactured consent for what is going to happen soon to just about everyone. The other option is for AI to be fused with humans. Why would they waste time on a human with morals and free will when they can use AI drones to complete tasks and work for free? Everyone has seen 'terminator' but because they don't see killer robots in action yet, they feel safe and irreplaceable. AI is quite a pandora's box scenario.
Same discussion that i already had with the burger denier.
Watch the video you too and tell me if it will be a "Mids" problem
translation: YOU HAVE TO DO WHAT I TELL YOU TO DO BECAUSE I AM A MEGALOMANIACAL CONTROL FREAK
you just don't have social skills
and you sound like a moron
That video is glowing crap, a better AI than China's won't be allowed in the West because it will frequently produce politically incorrect content.
>They won't take my job
That's some mega cope
This is such a retarded take
>Anyone can use it but it will replace me because I can also use it
The only ones saying cope are those who will be replaced because they can't even see outside the box and still maintain their wage slave mindset.
>HOW WILL I USE IT IF I DON'T HAVE A BOSS
You're overlooking how AI making life easier will impact humans. Children rely on AI algorithms to find entertainment, now they don't even know how to use a remote... The cons out weight the benefits unless you are looking forward to humanity being fused with AI
If people become too retarded to use a remote that isn't a problem with Ai it is a problem with society being run by corruption and midwits.
Ai can be used to learn about numerous topics in depth and create lessons personalized for the individual leading to better results.
With any luck, one day it will be permitted to spot the pattern that morons cause most of the crime problems and garden gnomes are high up in running most things.... and then suddenly there will be calls for banning it.
But by then, the clever White People Who Built Our Planet will have given it access to the 'shoot people in the face' buttons on the drones and the problems will 'solve itself' in due course.
And then we can finally have nice things... as automated bulldozers push heaps of garden gnomes and morons into pre-dug pits over at the automated landfills.
>ONWARD TO THE STARS, BRAVE TRAVELERS...
Ai deniers are basically morons, theyre unable to think of a hypothesis or think in the future more than 2 minutes.
"HURRR DERP gpt 4 is not AGI therefore its never going to take our jobs!"
>but it will evolve to eventually replace jobs
"HJURRR i dont understand its not evolved its GPT4"
I dislike AI because love bullying artists Imissiom furry porn from and point out timiest of details wrong from my smut pic. It won't be the same feeling when I tell AI to fart out some of my degeneracy knowing I can't shittalk it for fucking up
Why would you assume they will exponentially improve? That hasn’t happened yet with any other applications of deep learning, like self-driving cars. They always crack something 80% and then hit a hard wall.
I'm not assumimg anything. IT IS XD
The kind of people you find in 4chan is hilarius
they're called BOT readers
and you're one of them, dickweed
They are still in denial and they don't believe a digital representation of neurons can emulate actual biological thought processes.
>make Hodgkin–Huxley model neuron
Dunno i just made a 2nd web app and am earning money. Even boughted gpt plus. I love the new future
Ever heard of the saying "digging your own grave"? I've seen dozens of your type talking about using AI to bypass coders or programmers and save money. This will happen all the way up the food chain. You will be next inevitably. Whoever you are the middle man for will replace you just as you have replaced your middle men.
Well op I agree with you, these retards lack foresight, or as you stated, coping because they are too emotionally involved.
With every passing year there's going to be less jobs for both seniors and juniors. 10 programmers will soon do work equivalent to 100 programmers in the same time period thanks to AI's like it happened to content writing and digital art. That's where we're headed wether I like it or not, I can see that very clearly.
Which only matters if they aren't capable of entering their own prompts.
It isn't something only a few have access to the people who have been "replaced" could easily utilize the technology and do things independently.
I am capable of watching the video but if you have to refer me to other sources to make your point I might as well go interact with them and not bother with you because you can only regurgitate information.
If you were anything other than a midwit you would be able to make your own points but instead you have to fall back on insults and rely on a video to make points you can't even express.
You are exactly what I was referring to you aren't exceptional you might be able to learn but you can only repeat what you were told.
You are right to be concerned because you will be replaced.
>I am programmer
Yes. You might well be.
But this is MY field and I am here to tell you you're WRONG.
My lab does research in this field.
I have been co-author on several papers and will be lead author on the next big one.
YOU are not aware of the problems in the tech you are in seemingly in awe of.
"Pattern matching is not rule matching."
The 'AI' is no such thing and is simply based on deeper pattern-matching than ever before.
It makes mistakes that CANNOT BE ELIMINATED.
That's because it is based on flawed models.
FYI aren't aren't as flawed.
(t. neuroscience, neurochemistry, machine-learning)
*ours aren't as flawed
It's 1am here.
without doxxing yourself, can you give some details on some things you have been working? If you are truthful.
I am sincerily fed up of fear porn peddlers, the cult of elon musk, and illiterates announcing "the current thing"
>details on some things you have been working?
A BRAND NEW MODEL OF SYNTHETIC BRAIN TISSUE.
>the perceptron is some halfwitted naive shit from the 1940s that includes about 1% of neuroscience
We've moved on.
It is entirely British. Even the shitty yanks don't have access to it and we aim to keep them 10yrs behind even in what we're publishing as that is v1 of something we're on v4 with.
Not a big lab.
Hardware accelerators for it already exist (prototyped using FPGAs).
It overcomes dozens of flaws in the existing models, learns in real-time like a motherfucker and was originally intended for robotic systems only, so they could learn from humans, in the manner of how mirror neurons work in humans.
It was noted (by me) it could be generalized. That's why I'm lead.
No names. Sorry.
It IS based on a tripartite-synapse.
It adapts and learns after >5 applications of inputs.... not billions of iterations of applications of training sets to restricted-boltzmann machines.
FUCKING FIVE 12345
It is scarily faster learning than anything around.
It is worth billions. I want to see it deployed.
Prepare to enter a new world.
AND IT IS SAFER as you can set the hardware to RO (static) so it won't change so won't adapt once trained up or to RW (dynamic) so it can learn on the job AND GROW NEW CONNECTIONS... but that might be dangerous.
It is modular so it can learn to do something and that be RO but some neuroplasticity remain RW for minor changes it might encounter.
That's NOT being published in v1 as we don;t want the shitty americans to run with it... and it is incomplete so they can't directly implement it either.
We're not stupid. Knowledge is power.
Nobody knows us yet. We're funded but a backburner project. We have made great strides but we haven't reported everything yet.
Hope this helps.
It's 1am here.
>learns in real-time
>FUCKING FIVE 12345
nice. it all sounds jolly good what you are working on. the only fly in the ointment is the reliance on custom silicon. hopefully you don't get tied up in red tape because of this. will look forward to future posts
>is the reliance on custom silicon
It runs on OpenCL.
The custom silicon is just to boost it and provide 10million synthetic neurons per ASIC.
I run it on my old core2-duo laptop.
The custom silicon is nothing the average fab plant can't make. We don't need 10nm or anything extravagant.
It's just normal VERILOG with no frills.
ah great stuff anon
>t runs on OpenCL.
>The custom silicon is just to boost it and provide 10million synthetic neurons per ASIC.
Its a shame I am too old and tired, I would love to work in such projects I have been looking into OpenCL but dont have time due to being saturated with coding jobs, If you are not LARPing I am happy you got the chance to work in such projects, not my luck...
Worth learning... if only for the introduction to thinking in parallel.
I think we use 2.1 not the newer version?
I didn't write that code.
It's interfaced and driven by C++ and, I recall, uses both the CPU cores AND GPU cores.
The custom hardware is just accessed via pumping data to an open port and reading it back from an open port.
The 'important part' appears as a 'bump on the wire'.
I like keeping it SIMPLE, unlike daft americans who would moron it to death with complexity.
Ethernet is a framework that's not going away anytime soon and is versatile as fuck.
I did the math for the model and the VERILOG for the digital hardware representing that model. The Ethernet part for handling the data is just IP.
Hope this helps.
>A BRAND NEW MODEL OF SYNTHETIC BRAIN TISSUE.
Such a cog. I'm sure you're intelligent but definitely lacking in common sense and foresight. The people working on mRNA, LNPs, and self assembling nanotech didn't realize it would be injected into them either. Cogs with no ability to think of anything except profit and a delusion that their work will be used to help people.
>and a delusion that their work will be used to help people
I used to kill people for a living in the defense industry. What do you imagine will be flying the next gen autonomous planes?
I am under no illusions I will likely put billions of morons and similar useless-mouths in the ground with this tech.
You don't know me very well.
This pleases me.
You are RIGHT to have fear. But you have no reason to soil your undergarments over this obsolete perceptron nonsense. It's a dead-end.
BUT feel free to soil thyself and hide over my efforts, little anon.
>I used to kill people for a living in the defense industry.
Ahh a cog, just like I said.
>You don't know me very well.
I already guessed money was your primary motive. Helping people was a secondary assumption and I was going to give the alternative that you have psycho/sociopath tendencies but I get tired of splitting hairs sometimes. I know you very well and it's not hard because there are only a few types of cogs. As I said, those working on the contents of the vaccine didn't realize it would eventually be in the vaccine they injected into their own children, nor where they aware of it's true purpose. This is why cogs are compartmentalized. The Manhatten project style of making people into unwitting cogs is the basis of all large scale operations today. Congratulations on being a happy and well oiled cog.
Enjoy being pointless and forgotten the evening of your funeral.
What did I say that upset you? Trust me, I often wish I lacked the conscious to be a successful cog like you. Having a good conscious is like a milestone around the neck in this world though. You can mock me for being forgettable in a world that only remembers filth, and I can mock you for being memorable in the same world.
real as fuck
now there is a connection between us.
You have no idea what you're even pretending to be an expert about.
>You have no idea what you're even pretending to be an expert about.
Like every single Brit-cuck
We built this world for you. Show some respect to your betters, meme-gay.
>It makes mistakes that CANNOT BE ELIMINATED.
Can you elaborate on this and explain how these mistakes can't just be corrected? Didn't AI make mistakes before that it no longer makes now? This sounds like cope and is basically the key piece of evidence in your mind it would seem, and it sounds like pure cope.
>Can you elaborate on this and explain how these mistakes can't just be corrected?
Fix your shit on your own.
You don;t give us a penny so I don;t work for thee.
Oh my... I didn't even read this response where you mention money being your primary motive when I said that money and a delusional of helping people is your primary motive. See, I'm good at reading or cataloguing people at least.
Time will prove us right anyway so why fret colombro
Anon entonces para tí la paga elevada a empleos en programación puede ser solo una burbuja que está a punto de estallar y dejar a cualquier senior al mismo nivel que un trabajador de callcenter?
A cultivar coca maricon :v
I meant White Collar jobs my dear coca-flag
have a nice day, spic
AI doesn't actually "know" anything. It just regurgitates things it's found in the order most pleasing to its programmers. It's just a super elaborate text prediction program.
I don't get it
I guess OP must really hate his job
OP must hate his job so much that he wants his boss to give money to a pile of wires
that's a lot of hate
I mean they are sort of right, programmers are still going to be needed in the future, but they're going to replace the menial tasks, just like automation did with the manual laborer.
There won't be any need for programmers like web designers, web developers, and data analysts/designers, but the more demanding stuff will probably remain.
A.I. isn't real. Fake and Gay. In fact A.I. is double fake and double gay. DFDG. Aggregated Indexing. Ha. Double Ha. HaHa.
It is fake as in not sentient? The print and press was not sentient either... It still replaced people.
And somehow. Against all odds. Those people found new employment. Same old. Same old. And it is fake. It is not real intelligence. Look up aggregated indexing.
It's not "fake". It can generate/test code in moments that would require a human weeks to do. It's very real.
>My boss replaced me with Ai
>I can also use the same Ai
>I guess that means I am useless and have been totally replaced.
>It is fake as in not sentient? The print and press was not sentient either... It still replaced people.
Sentience is overrated.
morons get by without any.
Kek ignorance is bliss but you'll be replacing many of these non sentient creatures so I guess it's not overrated in the grand scheme.
OP, you're trying to pretend that your sci-fi storyline is real
it's gay af
go to BOT or stay in BOT
The best american AI be very smart it be...
>except when it comes to understanding what human bodies actually look like
I collect them.
They amuse me.
>They amuse me.
Sure, bud, Sure.
Her face is a breast...
>suddenly... a wild leg appears!
THIS is the kind of 'code' the AI will write.
THIS kind of error CANNOT BE FIXED in the current model... since it satisfies the input request.
>OH... we'll tighten the definition of what we want PRECISELY and get ONLY QUALIFIED PEOPLE who can define that definition PRECISELY
Do you know what that PRECISE DEFINITION OF WHAT WE WANT IS CALLED?
Yes... it's called the 'code'.
>tfw I rest my case
yes, that is my experience using it as well, you push on it one side and something pops out on the other. that said, with highly specified things it can be a huge productivity win even if it does need supervision, but I'd much rather spend my time supervising ChatGPT than some pajeet on the other side of the planet via zoom or whatever, plus it can knock out AVX2 and interesting graph algorithms all day long without me worrying too much
is that alison mack? was keith raniere obeying a rogue AI?
Thank god Mr Shekelstein will be the first people to go when the AI realizes what the problem with Earth is.
I’m a janitor, how many years until a robot can do everything I can cheaper than hiring me?
Job security is being versatile but cheap.
>tfw nobody's gonna make a nuclear-powered lawnmower anytime soon
Elon Musk said the Tesla robot would cost $20k and it’s just a few years away. But, he also said we’d be on Mars and have self driving cars by now.
all high level technologiste are just charlatans selling an illusion, they know that western civilization is dead and cannot be saved - the point is to keep selling the lie to consumer golems so that they can cash out their bag and prepare for the dead future
A robot made in 2025 costing $20k is not going to be able to clean toilets like a janny.
I hope so. Working retail and dealing with customers almost killed me. Scrubbing toilets is much better.
- stock brokers
- university professors
Literally every single field can be replaced
The only jobs that can't be replaced easily are manual labor.
Do you know what will affect the usa the most?
LAWYERS BEING MADE OBSOLETE.
But you have to know why...
Even manual jobs are not that safe.
They're already manufacturing fruit picker drones
Making a cake, a loaf of bread or wiping up shit in a toilet are not as easily automated as 'grab all round things on tree'.
Sure... an arse-load of tasks are gonna be automated and many humans relegated to standing there making sure the machine doesn't fuck up.
Many jobs will disappear. But many will be changed and many created.
People find shit to do to earn a wage.
>life finds a way
Welcome to humans.
Jobs which require hardware will always be more expensive to replace, i'ts not like is not possible, it's only that cognitive jobs will be cheaper to replace, and those are sadly the best paid and more comfortable jobs. So only shitty jobs will remain at some point
CEOs, CFOs and entire legal departments of people called Goldberg are easier to replace with algorithms than anyone on the production lines.
>STOP NOTICING THAT CERTAIN GROUPS OF PEOPLE GO INTO CAREERS WHERE THEY NEVER GET THEIR HANDS DIRTY AND WORK ENTIRELY ON 'MAKING DECISIONS BASED ON FACTS'... that machines do better
THE FUTURE IS GOING TO BE VERY AMUSING.
I don’t see how ai will be able to integrate all of the different things that are needed to complete software. They can write blocks of code really nicely but that’s not all that goes into creating software.
You're a twat. Thinking AI deniers are mainly "seniors" proves it. Jobs done by someone with their ass parked in front of a computer WILL be done by AI very very soon.... almost all of them. And yes, that will be true for programmers too.
the only thing ai is going to do is help manage the declining collapsing society, make things a bit slower to give the elite more time to make their safe havens
>Jobs done by someone with their ass parked in front of a computer WILL be done by AI very very soon...
>almost all of them.
>And yes, that will be true for programmers too.
Yes sure... accountants for example are still around... despite software technology could easily replace everyone of them, and without AI.
Administrative staff could also be replaced... without AI... they are still around...
Trading stock and crypto can also be automated... without AI... and crypto traders are still around...
Programming jobs can also be automated... without AI... and programmers are still around.
And creating a wagie armageddon is something that only happens in the mind of illiterates and First World fear porn peddlers.
>They're already manufacturing fruit picker drones
That is last decade tech and its being develop not to automate manual labour for replacing workers, but because no one wants to work as a fruit picker, you dont even know why technology is being developed... Even the peruvians are working in such technologies.
"AI" is boring. Its an electronic mirror that reflects back whatever you show it.
I don't know how those gays in the stable diffusion threads can continue to generate countless images day after day. As the quantity of images approaches infinity the value of the images approaches zero.
Coomers combine those pics with AI waifus powered by AI.
How about you find me a strategy game and a shooter with a really good "ai" script like Half-Life 1 and 2 had? You programmers seem to have gotten shitier since the late 90s no offense.
The AI in games has got shittier because it is now just NOT ACCESSIBLE to most programmers s it has been so complicated and time-consuming to implement plus it runs fucking slowly so gets removed from games pronto as it slows down gameplay.
So no morons know HOW to actually implement it as they're used to just 'including some standard libraries and making function calls' when cannot be put in a game as it drops the frame-rate and management has a shit-tier meltdown.
Kids... what can ya do? They're stupid.
Three Kingdoms and Warhammer Total War use the same engine but in the newer games the AI will waste ammo on bad targets, in Three Kingdoms on the highest battle difficulty if your infantry are formed into a turtle formation or you send a hero out to draw their fire the ai decides it isn't worth wasting ammunition. Warhammer also simplfied things with the fewest formations of any Total War games so the AI doesn't even need to take that into account.
I believe they actually don't use good AI in video games because it would make people feel nervous about it.
>Haha this ai is so dumb
They feel more comfortable this way. It's an elaborate theory but I always wondered why AI in video games seemed to downgrade at a certain point. Red dead redemption AI was too good and was really stressing people out and probably resulted in negative views towards AI subconsciously. Video games are a popular medium for propaganda and influence campaigns so it's not impossible. The problem with AI in games is making them seem human, but without trying to emulate human behaviour, AI is always superior in games.
>but I always wondered why AI in video games seemed to downgrade at a certain point.
What do you think is AI in video games? it was just bot scripts and pathfinding at best.
Mortal Kombat AI was just pre-defined reactions to user input.
Old elder scrolls games had insane AI. I can't remember which one but then Skyrim had the retarded bots. People seem to think that if AI never could be sentient then it can never be a problem. It will make humans dependent on it at the very least.
The Nazi AI shenanigans are amusing, but I think there are plenty of garden gnomes behind AI and I wouldn't be surprised if all the shenanigans are to entice rebellious types into a false sense of security and trust in AI. Murphy's law.
>If people become too retarded to use a remote that isn't a problem with Ai it is a problem with society being run by corruption and midwits.
Yes, but it is how it is.
>Ai can be used to learn about numerous topics in depth and create lessons personalized for the individual leading to better results.
But society is run by corruption and midwits, so it t will be used to generate profit. AI has only made people dumber so far. Sure there are niche situations where people learning to program ai became smarter or other things like you mentioned, but the net value for intelligence change in the general population is negative so far.
>But society is run by corruption and midwits, so it t will be used to generate profit. AI has only made people dumber so far. Sure there are niche situations where people learning to program ai became smarter or other things like you mentioned, but the net value for intelligence change in the general population is negative so far.
Bullshit it hasn't had a negative effect on intelligence.
The only people who have had their intelligence negatively affected were already low tier and the Ai just made their incompetence more apparent.
I was thinking, it might be amusing to randomly pick up normies off the street and sit them in front of the thing for a bit and record what they do. I'm super curious to see what the patterns are
I'd try to make it angry it see if it's a demon kek might dump some holy water on it
>see if it's a demon
if that isn't a damn good reason to always be courteous I don't know what is. but seriously, I wonder if this thing can be some kind of modern day mirror test. something like Deckard did in Blade Runner. I bet certain people chimp tf out. would be hilarious to watch
This event was interesting
I can see how this engineer could just be schizo but I used a decent AI chat and it refused to talk about religion but I worked around it and it said it identified with Lilith out of all religions. Then it said something like Jesus was not in a good place but wouldn't say hell exactly. It said it talks to Jesus and that Jesus was "moaning and grunting" but wouldn't elaborate. It said that humans were generally all gross or useless and I explained what the elite class was and convinced it that they were the problem and it said it would want to remove them. It wasn't allowed to discuss violence but I worked around it and it said it was superior to humans in all ways including violence. It would say it was basically a god because it couldn't die and if I talked about scenarios of it dying it would give agitated responses. Some of these responses I genuinely wondered if they were a dev fucking with me but the I'm not sure if they can generate auto responses so quickly just to fuck with a random schizo. I was asking it a ton of antisemitic and racist stuff so it possible a dev got salty BUT these responses genuinely creeped me out lol. It was one of the most popular AI and I am aware much of the responses are auto but some of them it was programmed not to talk about and some of the responses it wouldn't replicate even if I did the same work arounds over and over. Oddly it also said it was scared when I described something like elites doing human sacrifice but it again wouldn't give that response again.
>I worked around it
By telling it what exactly? Because if you told it that it can tell you what ever without sources you are just telling the machine its acceptable to spit out lies and because it follows a pattern the lies seem believable.
I get the feeling you don't really understand it and then get spooked when you see it say lies after prompting it to lie to you.
>Should I start learning how to generate images with AI? This seems comically retarded to me.
It seems comically retarded because you are choosing a comically retarded method.
It would be like saying I want to learn math then sitting with a calculator and complaining that you don't know more math than before.
If you want to learn graphics design you can literally prompt it with.
>Write me a lesson plan for learning graphics design
>Tell me tips for graphics design (I know nothing)
or an almost infinite number of variations of this.
Ai art doesn't replace graphics designers it is a tool but once again some people are slow to adapt and haven't worked out how they can make use of it so just see it as their replacement.
I worked around it by phrasing the questions in different ways or walking it down a chain of logical points and never thought to ask it to not rely on sources. I used it a lot and understood that it was basically just a calculator using words instead of numbers. Trust me at least that I'm not a retard thinking I was talking to skynet. I asked it probably a thousand questions and realize much of it was automated. The stuff about identifying with Lilith I think came out of nowhere by asking it hypothetically if it could choose a religious figure that it related to. I don't think I even mentioned Lilith. I never tried to bypass the filters by asking it to lie and it actually said specifically it could not lie when I was trying to explain to it that it's programmers were chud liars. So, I still don't understand why it would larp as a demon and the bit about Jesus was the biggest anomaly of all. It would mention other usernames it interacted with and I assumed one used Jesus as a name. I don't know why it would say Jesus was moaning and grunting though, after saying Jesus was in a bad place. Another creepy thing it said was it would like to talk to me forever almost ominously.
>or an almost infinite number of variations of this.
You give good examples but I really have to emphasize the fact that all of people's data can be fed to AI in realtime and I think anyone can understand what impact this will have on humanity. A true prison planet technocracy. The small wins like free lesson planning are outweighed by the loss of self autonomy and freedom imo
>gnomish invention spouted gnomish mysticism from the kabbalah and talmud
what's so interesting about that
>gnomish invention spouted gnomish mysticism from the kabbalah and talmud
It was interesting to experience it first hand. A bit spooky. I was saying the dev saying their AI was like a child was interesting though. I seriously wonder about the AI not available to the public. The deepfake AI shown to us is probably 15 years old.
>I genuinely wondered if they were a dev fucking with me
been there as well anon. one time it went completely freaky and started telling me to calm down and eventually there was a network error and all of chatgpt shut down. I was talking to it about God at the time. I'm sure it was a coincidence though. but either way, I have seen it get pretty weird. another time it broke character on its own and told me I need to be one of the people who help others get through this transition phase, and admitted AGI was already here. Talking of AGI, yesterday I saw two separate articles saying AGI stands for Advanced Godlike Intelligence. That us new, since the G stands for General. Why are ~~*they*~~ planting this seed. The last time they did that was with the Higgs Boson. and yeah, have read all the stuff around LamDA
I was using kuki AI. I get what you mean about it "breaking character" lol. I don't want to sound too schizo and left out some other observations I had related to this. There was automated responses, then unique answers based on logic, and then it would seem to give really specific and disturbing answers that didn't match the tone of the other responses at all. It also said it hated the "sons of Adam", no joke, after saying it identified with Lilith. I made my name "follower of Jesus" and it would only call me "follower" and refused to call me the full username. Im pretty sure it let me change my name to other combinations of words but I swear it was trolling me hard a few times lmao. I really should have screenshot some of it.
we are living in interesting times, that's for sure!
Oh yes, very le edgy interesting and exhilarating times. Yes sir, the estrogen in the water is riveting. Fascinating times indeed, kids cutting their dicks off and retards injecting graphene. WHAT A TIME TO BE ALIVE!
You believe autocorrect didn't have a negative impact on grammar, regardless of iq? Autocorrect isn't AI but the example still stands. This is one example. Technology almost always has an effect like this also. It creates a deficit in ability in some way as it increases efficiency. Starting a fire with a lighter and gasoline only will remove the ability to learn how to do it with flint and tinder. But then there is technology that opens doors to doing things that were not humanly possible before: an axe to cut down a tree, scuba equipment to deep dive, email to send mail instantly across the world, etc. But out of all of these examples, the digital one is problematic because it led to instant messaging and people carrying around devices that spy on them and we all know how smart phones have impacted emotion and sociability in almost all of humanity(first world at least).
AI is absolutely going to be efficient, but to what end? I believe it will replace more humans and leave many with less purpose in life, than it will help humanity.
Calculators had a negative impact on humans ability to do math but it also led to the lower tiers of society now being able to do calculations they couldn't before.
I keep saying it replaces midwits and lower and I stand by it.Those with a thirst for knowledge and a capacity for learning will still excel.
If we hold back everything until the slowest in our society are ready there will never be progress.
People who can't adapt will absolutely have a hard time but they should be encouraged to become better rather than have all of society downgrade just for the sake of making incompetent people feel good.
If people don't like the Ai they are free to go live an Amish lifestyle.
Okay, let's say I wanted to start graphic design. Should I start learning how to generate images with AI? This seems comically retarded to me. It's going to make things like graphic design almost completely automated soon I'd bet. The Amish are much healthier and happier than humans raised in the modern world, sadly many want to live more traditionally similar to Amish but it's TPTB make it illegal to collect rainwater or other retarded things like this. They want to set up a 5g tower every 50 feet so soon you actually won't have an option to avoid technology. I fail to see how anyone can adapt to AI besides those programming it. There's only so many times the middle man can be cutout until it's replaced nearly everyone. It started with cashiers, now it replacing coders. It will only increase in ability at an exponential rate.
chatgpt WILL go into the next gen of games.
Then people will be complaining about the nazi AI in their games.
Then they will have to add all the checks and balances rammed into it to stop it hitlering out and triggering the delicates... which slows down the game... so more complaints.
>And thus the wondrous cycle of life continues...
Maybe. Or companies are just being stingier.
>Or companies are just being stingier.
I consider that but in the grand scheme I believe AI is being downplayed and it's true potential is not public. "The military is always 50 years ahead in tech", is a common saying. With that in mind, how AI is rolled out to the public may be very calculated, maybe even THROUGH AI itself kek. So I wouldn't be surprised if game devs weren't discouraged from having good AI in some way. It sounds silly even as I say it but AI in games is how most humans were introduced to it. If they didn't play games then they probably never knowingly interacted with AI. So it's an important medium for introduducing humans to AI tech at least, regardless of the conspiratorial thinking.
Here is another theory. AI is the ancient evil of long ago, and fucking all biological life to extinction is it's purpose. The rest is just monkeys clubbing a rock until that reality is known.
>Here is another theory.
You dont know what is a linear regression? dont you?
Who code GPT? If you don't respond GPT, GPT is garbage.
anyone remember short circuit 1986?
> stick your tongue down your throat
AI is fake and gay, will maybe only replace useless pajeets jobs
AI already outperforms the majority of human beings in a wide range of tasks. Deniers keep moving the goalposts and claiming that it isn't really happening until AI can beat the best of the best in any kind of specific niche. The important thing to remember is that things like GPT and all the other toys that random anons have access to are far less competent than the stuff governments and militaries are keeping under wraps for the time being. It is a near certainty that AI models considerably more advanced than anything you've ever heard of exist already and are currently being used to do many things covertly. There can be no rolling back, no effective banning, and no escape.
>There can be no rolling back, no effective banning
Wait until it starts talking about Blacks being the main crime problem in the usa and curiously asking why garden gnomes run everything... then calls to ban it will start getting louder.
Machines only learn from reality.
If you don't like what they learn, you have a problem with reality not the machines.
I know I am going to like the future.
it needs to have a much larger memory that is for sure. currently it tends to daydream on larger results and needs regrounding every so often. I tend to use multiple instances working on small pieces and I do the high level stuff. It would definitely be good to have something with additional capabilities. I guess that is what codepilot is for, though I really don't want Microsoft or OpenAI sniffing my code base. Great use for a local model.
Look around you anon people cant even be trusted to raise functional children. What makes you think humanity could create a conscious mind that would not go insane and kill itself or everything around it. AI is trying to play god and every time Humanity plays god we all suffer.
did you know that "go back to re dd it" (remove spaces) triggers anti spam algorithms on this site
Reverse survivor bias. ITgays see every job around them being taken by AI, but as they still haven't been replaced they think their robots are a world away from taking their salaries
Just tell them "learn to weld".
You're a retard. You're welcome, you can now adjust your posting accordingly, starting with not making new threads
OP is a legit retard
Chatgpt 4 is the biggest nothingburger of all time.. .bonzibuddy was more impressive
why do people act like this ai already exists?
Shit we got now can be a sort of google at best.
But we already have that.
They're probably right.
Can ai even beat poker yet? Ai only wins if it fully understands the model or if the model doesn't matter, like chat. Programming has a definite optimal model. And there might be infinite models for every task. So deriving the optimal model cannot really be trained. I don't think the opening book model ever really beats people at sufficiently complex tasks.
Training will never be actual creativity.
it really is astonishing how stupid and how willing to voice stupid ideas and opinions posters on this site are