The Dead Internet Theory

[awoo]

I don't have proof any of you are real people either. I've seen some posters that not only have thousands of posts, but they also post long paragraphs about whatever is really going in in CWC's mind or how the entire modern world is a orchestrated fraud. The concentrated autism is like that of a bot.

 

[Mr. Hedgehog]

I'm impressed that nobody's mentioned the obvious: Presenting a fake consensus, showing that everyone's in agreement with a certain set of ideas, things the PR China is notorious for. Debating you isn't necessary. They could just a) hire actual people or b) bots to repeat similar things to APPEAR as if they're overwhelmingly popular. Normies are less likely to express themselves, or even look into "dangerous" ideas, as long as a sufficient number of "people" agree on not looking into them. People who shat on Trump for the last 5 years are the kinds of people who want a clear repertoir of topics to talk about, in a shallow manner. The weather, the nothingburger, Trump, "climate change". If tomorrow every outlet and everyone (every bot and paid troll, that is) appeared to be talking about aliens, they'd talk about aliens too. It's devastating for them to consider having a backbone at the cost of losing whatever mediocre social standing they have among other idiots.

TL;DR: They can just convince you that they've got the numbers and if you're scared you'll shut up and follow suit.

 

[Stasi]

I've definitely seen this and assume its shilling products or trying to get ad clicks as I've noticed it in tech related things. Like articles clearly stitched together from canned paragraphs or responses on places like reddit being completely detached from the question (and not just because of autism) e.g. recommendations for very specific things getting a bunch of responses recommending things not at all relevant to op.

Don't really know what solution there is or even if there is one other than trying to keep active on smaller communities rather than sewers like twitter and reddit.

 

[Yotsubaaa]

I don't have proof any of you are real people either. I've seen some poste

downloadable font: rejected by sanitizer (font-family: "quattrocento" style:normal weight:400 stretch:100 src index:0)<div class=\"message-userContent lbContainer js-lbContainer \" data-lb-id=\"post-8250027\" data-lb-caption-desc=\"awoo · Feb 1, 2021 at 3:25 AM\"> <article class=\"message-body js-selectToQuote\"> <div class=\"bbWrapper\">

rs that not only have thousands of posts, but they also post long paragraphs about whatever is really going in in CWC's mind or how the entire modern world is a orchestrated fraud. The concentrated autism is like that of a bot.</div> <div class=\"js-selectToQuoteEnd\">&nbsp;</div>
</article>
</div>



</div>

<div class=\"bbWrapper\">

What? Of course we're re


al. What, do you really think we're all just bots, parsing raw HTML and responding to trigger words? That's crazy talk, bro. I think you're all real and you are all learned and have worthwhile opinions. GME and always remember that the entire modern world is an orchestrated fraud and nothing matters.


Uncaught ReferenceErrolementByClassName("button--icon-reply").addEventListener(function() {/

 

[Tealeaf]

The “Dead Internet” theory doesn’t go far enough in my opinion. Does anyone truly believe there are billions of people on Earth? It’s a few tens of millions at most, the rest is just GDP manipulation. It’s impossible to support that large populations.

 

[Haim Arlosoroff]

While browsing the Farms an interesting remark came to my attention, something about a 'Dead Internet' Theory. I went looking out for it and found the following /x/ thread (archive). TL;DR the theory consists of the assumption that bots are being utilized to derail discourse on 4chan and most likely other online communities. What's the goal of said machine? Who controls them?

I don't believe they have an ultimate goal, but are merely a tactic.

Just as we give psychiatric medicines to developing minds as a tactic despite the resulting dependency on being made to feel less. Just as we societally have been given lives of leisure and office work despite the health effects. Just as our starch-based metabolism is unmade processing simple sugars.

We live in the times the elite want, they can 'care' after the 3rd world with the money of the 1st world. Their financial systems are already in place, as is their favorite form of government which can be bribed or staffed by them at will.

I adopt the "They already won" theory to the rich, this is the world they want from their perch over it. Yes the details are all wrong where they cannot see and it is all based on lies, but that was always the best they could hope for. Even the Saudis have their Aryan girl doing unspeakable things to a Bangladeshi boy. All is well for them. Then they see 4chan.

"Fix it!" they cry to their lapdogs, the true middle class who serve them most directly. The owners of service industries so bizarre you can scarcely believe it. Then the minions do exactly what the CIA has always done. Destroy what can be hidden, fanaticize what can be later turned in public opinion, defund what can already be lied about, demoralize what can be made to pay attention, and disrupt what is too noticeable and rooted. Tactics, woodenly implemented like the forces at the fronts of WWI. Only to be shifted not when the manpower is used up, not that sort of war, but when the elite cannot ignore a snub or a uncontrolled disruption to their garden of a society. Then the intensity is upped and like with $GME it can be foolishly exposed.

Still though, the effect of suppression will kill the joy of things. 8Chan was always the cause of 4chan going to shit though, I think, because it caused a flavor of user to leave and the resulting website because dull and vapid. Boomers after the Trump thing, whatever that was I still cannot say, were another nail. Too many nails kills the Internet. Although advertising and money exchange professionalizing the internet, and nerd-face TV shows popularizing social media didn't help either. 2008 was the beginning of the end.

 

[Large]

It's not bots. It's people managing dozens of social media accounts, possibly through special software designed to increase their productivity (which is a possible source of those AWS logs that redditor observed).
Politicians and corporations pay them to shill.
Only relevant english article I could find is from The Guardian, but it's fairly informative so here's the link: https://www.theguardian.com/world/2...-reporter-reveals-life-in-a-polish-troll-farm

 

[Mr. Boogedy]

There's definitely bots on Twitter and Reddit. There are novelty accounts that are openly bots that can respond to people. I have no doubt that bots that are supposedly real people exist. The problem is that the characteristics of bot posts - posts consisting of set phrases rearranged in various ways, only responding with set phrases instead of addressing the post, ungrammatical or incoherent posting - also apply to real people. It can be tough to tell real humans from bots. But on the other hand, it's also cheap to pay third worlders to spread your message.

Bots pretending to be real women are used on dating sites to get men to buy memberships. The Ashley Madison leaks showed that at least 70,000 accounts were fake. These accounts could respond to messages from male users, encouraging them to purchase services. Some analysis suggested that the majority of female accounts were fake.

There was a history of bots posting on Usenet going back to the early 80s. As garakfan69 mentioned, Mark V. Shaney was a Markov chain program based on posts on and posting to net.singles created by employees of Bell Labs. Many people thought that Mark was a real, if odd, person. A Markov chain, to simplify, analyzes a corpus of text and creates a statistical model weighing the likelihood of certain words or phrases following another. This model can be used to generate new texts which are largely grammatical, and can be seemly meaningful for short texts, but large texts are disjointed or incoherent. Markov chains are still used, notably for Twitter *ebooks accounts.

Another famous one was a spambot called Serdar Argic, the bot would scan all of Usenet for mentions of Turkey or Armenia and post messages arguing that the Armenian genocide never happened and that actually it was the Armenians that had committed genocide against Turkey. The bot famously would respond to posts about the animal or food turkey.

Several years ago I decided to check out Usenet. Outside of piracy groups and a handful of legit boomers still posting in certain nerd groups, it was like a ghost city. I saw dozens and dozens of discussion groups filled with nothing but spambots. Months and months of posts from multiple spambots posting every day. I found it strange. Spam killed Usenet, and long after any actual posters had left, these bots were still there, posting spam that almost no human would actually see.

 

[teriyakiburns]

Twitter is mostly bots. Its data analytics services are basically useless because of this, and because the rest of the active accounts are a self-selected minority that skews young, western, stupid, and left.

 

[Governor Jeb Bush]

Maybe we are all glowniggas, speaking to other hired glowniggas?

 

[helloworld.exe]

This is why alot of people won't investigate anything that contradicts their belief system. Subconsciously they know what they'll discover. Consciously they're not aware of what their subconscious knows. Their intuition is flawed through social conditioning and it guides them away from the truth in order to protect the fragile soul that's lost and confused inside. There is no heaven and hell after death, throw that lie away. Hate to be the bearer of bad news but we're living in hell. Never fear, discover then accept the truth and "heaven" is near.

 

[AmpleApricots]

I enjoyed a lot playing around with the GPT3 model you get relatively easy access to via the "game" aidungeon (really isn't a game and the devs of that thing are webshits that realllly don't understand this sort of technology) and while it's sometimes downright uncanny and surprisingly clever measured in human conversational qualities it has the attention span of a 12 year old with ADD high on sugar combined with the memory capacity of an 98 year old with late stage dementia and getting it to stay on topic is like herding cats. I do see the potential in that technology to be one day at a point where it'll be able to post on something like twitter, even naturally interacting with other people without you being able to tell if it's a person or not but that technology is not quite here yet and if it is, it's way too expensive and error-prone to be practical. I think most "bots" are just a bunch of formerly starving linguists or other low level college/university educated government-employed workers, probably operating dedicated tools to control tons of accounts in a convinient way, the infrastructure for that being done by other people of the same department. Some fancy UI where they basically get thrown into a discussion probably determined as interesting by some algorithm, have to write something fitting in line with whatever is dictated (maybe even being able to answer to their own posts with different accounts and such stuff) and then click "next" for the next one. A lot cheaper and works a lot better. Could totally see such "bot farms" happen in China and Russia. Purely automatically if you want to be above the level of P3N1S enLArgMenT spam though? Nah.

 

[jimsterlingspronoun]

the usual subjects jews, marxists, nazis, qanon, white supremcists and the amish

 

[The Real SVP]

I think most "bots" are just a bunch of formerly starving linguists or other low level college/university educated government-employed workers, probably operating dedicated tools to control tons of accounts in a convinient way, the infrastructure for that being done by other people of the same department. Some fancy UI where they basically get thrown into a discussion probably determined as interesting by some algorithm, have to write something fitting in line with whatever is dictated (maybe even being able to answer to their own posts with different accounts and such stuff) and then click "next" for the next one.

As @Large said, and my own comment hinted at, it's call center agents pushing buttons. There is a database and some AI systems in the background, but most "AI" systems that can hold or contribute to a conversation are people choosing from premade text blocks with some variables included.
There are also large operations, mostly in China, the US, and Germany, without any sophisticated back end. Just groups of human agents, each playing the roles of multiple online personas. They tend to be in a higher pay grade and are expected to take care of organizing the data they need by themselves. Usually they use a couple of text files to keep track of their past posts, and to contain text macros for fast replies.

 

[L50LasPak]

It's not bots. It's people managing dozens of social media accounts, possibly through special software designed to increase their productivity (which is a possible source of those AWS logs that redditor observed).
Politicians and corporations pay them to shill.
Only relevant english article I could find is from The Guardian, but it's fairly informative so here's the link: https://www.theguardian.com/world/2...-reporter-reveals-life-in-a-polish-troll-farm

This software does exist and is in widespread circulation. Though I think its working in tandemn with automated software. There might also be some cross-connectivity between the two softwares so a human can momentarily take over and defeat a captcha for a waiting list of a hundred robots or something.

 

[Liber Pater]

I remember back in 2016, after the election of trump, a document was leaked that discussed how Share Blue and other liberal think tanks were going to divide Trumps supporter base in light of his success. Among them were things like saying he abandoned them, discrediting all his accomplishment and using all of his compromises as weaknesses, and saying that Trump didn’t actually represent modern right wing ideology. There were other tactics in there too, like banning conservative hubs and disenfranchising websites using financial tools. It’s scary how accurate that document was in retrospect. I remember laughing at it, saying to myself “this is absurd. We’ve won. The election of trump is going to start a new age of conservatism”. Looks like we were playing checkers while they were playing chess

No, we were playing chess and they just took a shit on the board. And yet we kept playing.
Only one party got tricked/browbeaten into thinking there were rules.

 

[stares at error messages]

<https://forum.agoraroad.com/index.p...net-theory-most-of-the-internet-is-fake.3011/>
This reads like Tatsuhiko Takimoto Welcome to the N.H.K. Even the metal gear sounds like Cereal Experiment Lain. Boo! the memes are alive. Makes me think about how much of society is dysfunctional and keeps up lies about its self. Like the cultural equivalent of low-self-esteem. Highschoolers are said to be retards even though they really just want to grow up and take-on adult responsibilities. Arguably the exact opposite of being naive foolhardy. Adults keep meming the lies and they just don't work anymore. Women are supposed to be the image of kindness and virtue and there have been a handful of cases where teenaged women and older have bullied their peers into suicide or actually killed each-other. But in these cases they are only put away for 20 years because the jury wants to fuck them. Most women in and past 20s are collapsed narcissist who sit back and heckle at the world; you're at fault when you're miss treated and you are at fault when you make a mistake, so you really have no agency at all. Porn will never leave me and sue me for child support. Then they grow up into being afraid of their own shadows and can't even open their own jarfoods.

It's more comfortable to believe that the memes are alive then it is to think that society is out of its depths entirely and will fall apart in the next few decades. College and Education TM doesn't work anymore. (if it ever did) Xers who have their jobs now will have them for the next 20 years and deprive their kids of a future in the process. Identity politics is the new religion because America never had a religion other then praise me. Opinions were too expense so we'll just have echo-chamber-hug-boxes. And advertisers randomly try something else to help their products sell more, like any of the thing they've tired ever worked at all. No they don't work and they're more happy to have 1000 more Theranos. Women, Minorities, Environment, Blacks, Trannies, Gay Hatred, Gay Pride... Just ruling out all the options. Like a virus going though permutations. It's evolution. Just ask a Giraffe why its has brain aneurysms pushing blood up that long neck of his. Death and extinction are natural parts of evolution. Europe is now a Muslim caliphate and North America looks more and more like South Africa every day. It's all dystopian. It would be nicer to believe in a plan or a grand plan that only you have discovered because the great being conspiracy is keeping it secrete and now you've cracked the code. How big you must think of your self to be. It's all been a dream. It's all been a AI computer program and not an alarm you're tied of hearing.

 

[teriyakiburns]

It's all been a AI computer program and not an alarm you're tied of hearing.

Why do I so strongly suspect that you're posting the output of GPT-3?

 

[Haim Arlosoroff]

Maybe we are all glowniggas, speaking to other hired glowniggas?

I mean, the alternative is to autism out in such a particular direction that it is unlikely to be programmable and still be relevant. Plus SS13 shows the limits of bots, what I would not give for a half-filled station of bots you couldn't tell were bots unless you conversed with them. Engineering could have competent setups, chemistry and robotics could be adequately staffed to treat players and through borgs allow respawns.

There, there is autism which is more likely a dumb-ass ape than Markov chain. Today.

 

[Temporal Refugee]

I find the likelihood of machine-assisted "troll armies" to be far higher. That is to say that, while I don't doubt laypeople such as us are unaware just how coherent/complex institutionally backed AI can be, I believe the more effective methodology at this point is most likely to be setting up very small and isolated (but well trained) teams with AI supplemented tools to derail, backslide, and manipulate public perception. Think Cleverbot vs someone on their phone using autocomplete.

A fully automated solution will likely shit the bed and shatter the illusion at some point, raising red flags among the targets (users) reading their comments, posts, etc at some point. Human involvement acts as a bulwark against that, maintaining the simulacrum. At the same time, it would be more cost effective and less prone to failure to design tools that offer up entire phrases, sentences, or even paragraphs for a person to tweak and ajdust before posting before they hit a hotkey combination or the like to switch over to another one of the 100 accounts they're obligated to express the same ideas from before EOD.

Hell, for that matter it would be trivial to also have a tool that records the posts being made by the entire team and warns them/disallows posting if the output to be submitted happens to be < x% dissimilar from one already made, with some form of heuristic learning to grow better and better at identifying synonymous phrases in memes and other forms of informal expression.

TLDR - it's not all bots, but anybody who takes things at face value in terms of the internet's population, or blindly follows the perceived consensuses without question, is naïve.