Thread 23119271

Newfag here but why is /bant/ the home of fishtank threads? Wouldn't that be more of a /soc/ thing?

they got moved here from wherever else they were shitting up

You are so lucky I am around and have a feeding complex.

You see baby Anone, the media boards have a little unspoken rule where if the series doesn't get a new episode/season with an arbitrary lapse of time, the threads are declared a general. The fishtank sepo mutants come from the /tv/ board. Those things were quickly exiled like jews because they are simply insufferable faggots with little impulse control and regard for general 4chan decorum. Not to say, those threads pop up from time to time on /tv/, I dont keep up with them. Simply understand that the board is not only an internation board (and a trash can for /pol/) but also a:
R A N D O M
O
A
N
D
O
M

board where anything goes. Even users here forget that. I guess Anones here get stuck on the '/' character becase raeding to hart.

Now lemme wipe your little fat baby face, Anone, you are drooling again ~~~

Generals shouldn't be on random boards change my mind.

>>23119353
Agreed, it’s annoying.
The only interesting thing to come out of fishtank generals was the drama involving Betty and her psycho stalker simp orbiters.

the fishtank threads are sophisticated "false engagement" AI talking to each other/themsleves/itself
it's just a moneymaking farm for sam hyde

dunno if it was always like that but it's very obvious
you can check out the russian meliorator software from a few years ago if you don't believe me, it was partly based on opensource stuff and it did the same exact things these threads do

>>23119326
it was /tv/ and I think they should go back!

>>23119409
IDK if I believe that, AI is VERY hard to train to act realistically human for a long period of time. Maybe it’s just me but I’ve been playing with LLM since they started becoming public and while im no AI expert by any means I have a very good feel for them due to the amount of hours I’ve spent playing with them. The hardest human factors to replicate with an AI or consistency and memory think of it like you’re an investigator doing a police interrogation but the person you’re talking with is lying and can’t really keep their story straight. It’s the same sort of feeling talking with an LLM that’s attempting to replicate a character/person. Sure they can keep up the act pretty good at lower context/low word count but as the conversation goes on or becomes more complex they’ll eventually start slipping up in was a human doesn’t normally do. Little cues here and there like incorrectly recalling events or changing opinions suddenly or reacting to incoming information out-of-character. LLMs seem to struggle in particular with numbers and human biology. These are all errors I’ve noticed with Chatbot/story writing LLMs usually only involving myself, the AI character, and sometimes another character. This is relatively speaking, a fairly low entropy environment. Now imagine that you are attempting to simulate a image board with multiple AI characters interacting not only to each other but events on a live show in real time, while adhering to 4chan format (something I doubt there’s a lot of training data on since I’ve never been able to successfully replicate with an LLM).
I don’t think it’s impossible by any stretch with some of these new unreleased models and they’re touted capabilities, but I thinks it’s highly unlikely those resources would be waste on something as silly as Sam Hyde’s Big Brother clone.

>>23119484
>Sure they can keep up the act pretty good at lower context/low word count but as the conversation goes on or becomes more complex they’ll eventually start slipping up in was a human doesn’t normally do. Little cues here and there like incorrectly recalling events or changing opinions suddenly or reacting to incoming information out-of-character.
i think this happens when i talk in those threads, but i haven't done it for very long
>>23119484
>This is relatively speaking, a fairly low entropy environment. Now imagine that you are attempting to simulate a image board with multiple AI characters interacting not only to each other but events on a live show in real time, while adhering to 4chan format (something I doubt there’s a lot of training data on since I’ve never been able to successfully replicate with an LLM).
the important real-time events could be transcribed/written and fed to the system by one paid guy, possibly
>I don’t think it’s impossible by any stretch with some of these new unreleased models and they’re touted capabilities, but I thinks it’s highly unlikely those resources would be waste on something as silly as Sam Hyde’s Big Brother clone.
it's not about new and unreleased models, or even LLMs
it's about neural networks in general
sentiment analysis programs have been around for over six years, same with computer vision
LLMs are a single parlor trick of neural networks, which are fairly versatile and useful in their place
everyone is so distracted by the consumer models and what nerfed LLMs can do
you should check out the youtube channel sentdex and look at some of the older videos, if you're interested
he also wrote a book on neural networks
all the modern "AI explained" videos are so full of slop
these are basically like cooking videos, it's one mildly autistic dude just building things and explaining his thought process
the guy who made the modified baritone AI for 2B2T was cool as well, that was also over 5 years ago

if you go through those older sentdex videos, even though he never actually talks about LLMs beyond reviewing github's copilot (which was sort of a predecessor to the modern stuff), it'll break alot of the magic/anticipation and you'll be able to see the limits of these things, predict what the companies will do & why, and see how neural networks are useful in their place

there's also that meliorator software i mentioned, it does exactly what i'm suggesting the sam hyde threads are except it's meant to be used on twitter
the thing keeps track of several personalities/"souls" at once, and has them interact with one another & react to outside people commenting, in a way which moves some specific narrative forward

>>23119345
this is the only right answer. anything else said in this thread are made by retarded redditors and newfags that never lurked.

https://www.ic3.gov/CSA/2024/240709.pdf
is a good summary but it's kind of abstract
there's info/examples of it in action if you look around

>>23119533
i didn't mean to answer the OP question
it just aggrevates me that the fishtank threads are treated as organic or given any attention

>>23119484
>AI is VERY hard to train to act realistically human for a long period of time.
also, can i ask what you mean when you say "train"?
it has an older more specific meaning within neural networks, but i feel like when people say it they're talking about LLM prompts

and yeah i do agree it would be hard to pull any of this off using some consumer shit like chatGPT

>>23119520
Oh wow, this might be a little much for me right now I haven’t even started my introductory courses for programming yet. But I’ll keep this in my watchlist ‘cuz I’m definitely curious.

>>23119567
it is the best
>this might be a little much for me right now I haven’t even started my introductory courses for programming yet.
i felt the same way, something about his videos made it seem so straightforward
he does it all in python which is a good starting language, and has a short python course
>But I’ll keep this in my watchlist ‘cuz I’m definitely curious.
:D

i gotta go to colelge

his voice sounds like edward snowden's as well, which is relaxing

>>23119528
>there's also that meliorator software i mentioned, it does exactly what i'm suggesting the sam hyde threads are except it's meant to be used on twitter
the thing keeps track of several personalities/"souls" at once, and has them interact with one another & react to outside people commenting, in a way which moves some specific narrative forward
Once this technology becomes commonplace enough for scammers to use I think we’re gonna see a market crash. How are you supposed to sell product to shareholders if you can’t guarantee a product is being seen by people? Same thing with add clicks. How do we know those clicks are from real customers interested in a product and not bots?
Those are just some examples I can think of off the top of my head how AI could be used for market manipulation.

>>23119555
Aren’t neural networks like LLMs and image generators “trained” off datasets at their core? Like billions of references images/text that it learns from. At least that’s how early AI image gen worked to my knowledge.

>>23119615
>Aren’t neural networks like LLMs and image generators “trained” off datasets at their core? Like billions of references images/text that it learns from. At least that’s how early AI image gen worked to my knowledge.
yees, that's what training always meant as far as i knew
it seems like that might've changed with how commonly used these LLMs are now, idk though
the training phase is orders of magnitude more computationally intensive than actually running the finished model
the guy who made the baritone model i mentioned, he had to rent microsoft azure servers during the training phase, but the finished product (sophisticated bot that plays minecraft, basically) can run on consumer hardware
i think he intended to sell it to the highest bidder
but yeah, you could train an LLM based on gigabytes of reddit/stackexchange posts/interactions, for example
you can train a computer vision classification model in a similar way, say by giving it like 10,000 pictures of dogs/cats, each labelled as dog or cat
>Once this technology becomes commonplace enough for scammers to use I think we’re gonna see a market crash. How are you supposed to sell product to shareholders if you can’t guarantee a product is being seen by people? Same thing with add clicks. How do we know those clicks are from real customers interested in a product and not bots?
it's gonna be interesting to see how it all shakes out, i don't really know
but i definitely think the more interesting stuff is gonna come from chaining different types of neural networks and other programs together in ingenious ways, and possibly from breakthroughs in the matrix math these things are based on in the first place

i know things like chatGPT, they have a limited amount of "memory" and prompts/responses you can have during a single conversation
i'm not certain how that works, but i think i know
basically chatGPT is connected by billions of "neurons" connected in a web
you make a pass by feeding it a question/prompt, and then it spits something out
however, some neural networks are "recurrent neural networks", where each pass actually MODIFIES the neurons/weights, in a way which depends on the data you feed to it
i think this is how chatGPT is able to achieve a sort of "memory", where it can recall prompts you gave earlier
however, with each pass and modification, the thing becomes more schizophrenic and unpredictable, basically
the company probably sets the limit of questions per conversation WELL below the point where it would become schizophrenic

but yeah normally a neural network or LLM wouldn't have any sort of "memory",
you make one pass & it spits out an answer & that's it
nothing resembling memory beyond whatever it's trained on

>>23119666
>however, with each pass and modification, the thing becomes more schizophrenic and unpredictable, basically
The simple answer to this is mathematic entropy, https://en.wikipedia.org/wiki/Entropy_(information_theory)
This video is old but does really good job explaining the theory around AI and neural networks, though I think some of the terminology has changed it’s still the video I recommend people watch if they want a simple explanation for the theory.
https://www.youtube.com/watch?v=wgCuKTN8sX0&t=1146s

>>23119727
ty
i'd heard of entropy with respect to cryptography, but i didn't really spend much time understanding it
>IMG_8670.jpg
i need one of those chairs

stop wasting your time learning about ai it's all going to be wiped off the face of the earth within the next decade when the neoluddites commit a butlerian jihad

>>23119758
maybe
>humans develop technology
>humans develop AI
>AI enslaves humans
>solar flare disables AI
>humans construct pyramids to worship the sun god
>dark age happens
>cycle repeats

i heard the global power grid could be disabled with about six well-placed nuclear blasts high in the atmosphere, i guess they produce EMP bursts
i hear the power grid would be very difficult to fix if the whole thing shut off at once
it's funny how america has three power grids; west coast, east coast, and texas

>>23119271 (OP)
After season 1 ended these freaks were still obsessed with the show and kept making generals on /tv/
The mods decided to dump the /ftl/ threads here instead of /trash/
Of the times I've lurked the /ftl/ threads there's been a couple people who believe sams crew is keeping the /ftl/ threads alive, though those could just be schizos.

>>23119758
It took me a minute to realize this was a dune reference lol.

i thought it was warhammer at first
i even read dune & i still didn't remember

>>23119406
I think they're just simps dude

Btw the reason why they were able to station themselves here is that /bant/ was already a vulnerable place that wasn't much of a community anymore, and the posters no longer cared about it. It was already invaded by others like /qa/fags.

Rope yourselves, /bSCAT/. /ftl/ is here to stay. Look forward to more /ftl/ spinoff generals in the future.

>>23119845
i don't believe there's enough simps to steadily get 1800 forgettable replies per day across five different threads
i also think it's funny how the threads basically got kicked around different boards, and were just okay with it
i guess you could ask them about it, but i'm guessing they wouldn't give any clear answers, even though they should be pissed about it
but if you're running an engagement farm, SOME engagement is better than none, so it makes sense you'd be okay with it

>>23119859
Flood the catalog so jannies are forced to put you down

/qa/ should've won with faggots like this >>23119859

4chan is a dead website. We might as well do as we please before it goes away for good.

>>23119901
bleh

>>23119901
If 4chan dies there are plenty of replacements waiting to fill the void.

>>23119907
pitou :D

i wonder if this is all disruption from the AI talk lol

>>23119907
Remember when 4chan was down for a couple of weeks after the hack? The userbase was so fragmented that didn't work out very well.

Also I haven't seen an ftl thread in ages
/\/[a-z]{3,}\//i

>>23119912
Why would it be that?

>>23119919
My bant thread on 4x2chan had like 700 replies though. It really wasn't that bad if you knew where to look.

>>23119931
>Why would it be that?
it's bad for people running engagement farms, if people are talking about them & becoming aware of them
it's also bad if they're exposed as inorganic threads, since 4chan staff might remove them
i may be schizo, but if someone DID have the ability to run those threads, they'd have the ability to easily disrupt any threads talking too much about it
there's been sentiment analysis programs for over 6 years, initially for automatically analyzing news and conversations and using them as stock trading signals

>>23119936
It’s organic bots, not some advanced network of bots talking to each other to advertise fish tank.

>>23119955
from the thumbnail i thought that was pitou facepalming impatiently/embarrassed, but i clicked & the eye spooked me lol
>It’s organic bots, not some advanced network of bots talking to each other to advertise fish tank.
it's not that advanced at this point, but yeah, i think you're right now, about this thread atleast
still, i think the threads are artificial

>>23119919
I went briefly to -that- website (the one responsible for the hack allegedly) for a while since they have a /qa/ board but it was to slow for my liking, everyone posted on /sōy/ and /pol/ so there was no real engagement and the other two boards moved so fast with the influx of refugees it was impossible to keep a thread going. Also posted a bit more on CC than normal but it still had the same issue with low PPH count.

>>23119862
>i don't believe there's enough simps to steadily get 1800 forgettable replies per day across five different threads
you think too highly of people

>>23119959
>about this thread atleast
I'm right about all of them
>i think the threads are artificial
Enough of the mentally stunted like it to keep it alive.

>>23119871
'tarded outliers don't define the entirety

>>23119785
Those threads are 100% shills.

>>23121050
I like fishtank and I've never spent a day, they're certainly way better than tripfags threads

(no offense Brock)

>>23121058
>iv never spent a day
What does this mean? Are u saying u never post in them or u never go a day without posting in them? Or something else entirely

>>23121058
it is ok

>>23121165
g'day