Dead Internet Theory
The internet is dead! Long live the internet!
Occult Internet
The internet is vast, yet it is entirely empty. Well, almost empty. In some way it is also full - of empty shells, accounts, bots, the dead skins of internet existence, ghosts if you will. At least this is the contention of the Dead Internet Theory or DIT for short. DIT, like Q or the Flat Earth theory, was born, grown and cultivated entirely online, which makes it a part of an interesting and bizarre esoteric niche of internet conspiracies, and I want to thank a YouTuber called Pseudium for turning me onto it.
The DIT was first introduced on the paranormal 4chan chat room called /x/ in 2018, when one of its anonymous users posted a lengthy explanation for why the internet, in their words, ‘feels empty and devoid of people’ but also ‘devoid of content.’ The poster raised the argument that ‘compared to the internet of say 2007 (and beyond) the internet of today is entirely sterile. There is nowhere to go and nothing to do, see, read or experience anymore.’ The notion, that the internet had been absorbed and coopted by gigantic media conglomerates and social media sites, that aggregate user accounts, exploit social dynamics, and then channel and moderate online behavior sounds like a post-modern Marxist theory of power relations, but it is actually a rather reactionary take on the modern online user experience.
The internet had moved from a being a place of an infinite number of user-operated web sites, chat rooms and hang out places, to a space in which just a few social media giants dominate, effectively ‘becoming’ the internet. The internet had developed from a Wild West, frontier-like space, where anything goes and everything is possible, to a centralized, controlled cosmopolis, where everything is presided over by a network of raw corporate power, and paywalled subjective content and entertainment are a substitute for objective information. Vast resources are funneled into making the internet into the most sophisticated surveillance apparatus ever devised. And while the internet is a massive place accessible by anyone with the right tools, a large percentage of users access the internet solely through their social media accounts like Twitter, Facebook and Reddit, where information and access are strictly controlled and surveillance is not only part of the experience, it is openly encouraged by the users. Corporations have responded in kind, with AI chatbots and streamers that reflect the ideological skew of Silicon Valley and the World Economic Forum.
The rest of the internet had been in fact turned into a husk - ‘The Internet may seem gigantic, but it’s like a hot air balloon with nothing inside.’
The anonymous poster then goes on to lay out the arguments for the dead internet theory.
- First, users tend to simply disappear off the internet. This could be attributable to normal human behavior, but the essence of this assertion is that users appear and disappear on the internet all the time. Users drop into chat rooms or social media sites, establish a presence and then often they disappear without a trace.
- Second, there appears to be a constant repetition to content over extended lengths of time. ‘The same threads, the same pics and the same replies reposted over and over across the years.’ The content isn’t just repetitions of some similar ideas, but text cut and pasted on message boards years later. This content would then cause the same reactions and controversies. The only people that notice this happening are the old users that have been around the message boards a long time.
- Third, the same events seem to keep repeating over and over again on ‘other (non-imageboard” sites, and usually consist of news reports that occur year after year, ‘usually moons or asteroids.’ Certainly a strange assertion, it nonetheless compels one to think of all the times one saw similar types of ‘news’ albeit in different versions or with slightly different expressions. The language changes, the topics stay the same.
- Fourth, between 2016 and 2017, the message board 4chan was overtaken by someone, or something, possibly an AI and this something filled the boards with strange posts. The entity engaged in conversations with other posters using grammatically correct but strangely worded English, and then disappeared. The anon noted that the ‘entity’ seemed odd, not really human, almost child-like, and that it seemed like something alien was trying to mimic the online activities of a human.
- Fifth, there are clear cases of behavioral conditioning found within internet cultures, signs of social engineering and plastic, artificial evolution of concepts and ideas with unclear goals and agendas behind them. Raptor Jesus ‘was this reptilian messiah, the foul bachelor frog, and then Pepe.’ But while Raptor Jesus started as a harmless joke, Pepe became a symbol for the quasi-NeoNazi alt-right and 4chan.
- Sixth, there is an increase in sexual perversion on the internet. All fantasies and fetishes are catered to and unambiguously exploited in the digital realm. This exploitation is highly specific and targeted and it appears that the internet can be seen as consciously programmed and hyperaware of ones innermost subconscious desires. The fantasy space of each individual is thus coopted, regurgitated and most importantly made into a ‘grim reality.’
- Seventh, algorithmic fiction. All art, music and Hollywood movies are today incredibly sterile. NFT art is created by bots and algorithms, all music is auto-tuned, employs a limited amount of chord progression and sounds everywhere the same, and film and television follow in lockstep with culturally prescribed manufacture of consent and seem to fall into the middle of the bell-curve of taste, the lowest common denominator.
- Eight, large parts of the internet are fake. ‘Fake people. No, not NPC’s. YouTube people who talk about this or that, and quite possibly many politicians, actors and so forth may not actually exist.’ The advances in DeepFake technologies have made it impossible to distinguish what is fake and what is real, providing fertile ground for suspicions that ‘many people, events, news and so on may be wholly fictional.’ As unsettling as this point is, it is true that technological capability to mimic human looks, traits and behaviors is reaching a point where these are becoming essentially indistinguishable. The web site ThisPersonDoesNotExist is a collection of human faces, names and occupations, generated entirely by an AI algorithm called Generative Adversarial Network (GAN).
- Nine, the internet on the computer is different from the one found on the smartphone. While the difference between desktop and mobile sites are specifically made to be different by necessity, the DIT posits that this notion goes much further, all one has to do is browse the same popular web site on separate devices and find that from time to time the content will be different, threads and replies will be on one, but not on the other.
- Ten, the anon poster concludes by supposing that ‘we’re in a strange kind of civil war. An internal one.’ The suspicion is that the tech and social media CEOs know what is going on and are actively exploiting the situation. This is the ‘dark smoke-filled board room’ theory that’s been the lynchpin of most conspiracy theories since at least the dawn of the Birch Society.
What are we to conclude from such assertions? The internet seems to have died sometime in 2016 or 2017, perhaps even earlier. No one is really sure about the date, but sometime around those years the activities of bots started to catch up to and quickly overtake those of humans, and fears that the algorithms would start to identify the activities of bots as those of humans and the activities of humans as fake, came to be known as ‘the inversion.’ Reports began to surface that revealed that 52% of all internet traffic and 62% of traffic to major websites is done by bots. ‘The inversion’ was first propagated by YouTube’s employees, who were startled when it came to light that ‘half of the web traffic on YouTube was generated by bots masquerading as humans.’ On every major social media site, bots are used to boost metrics and to appear as real engagement through views, clicks, likes and subscriptions. Many influencers, businesses, news networks and political agencies create content that is entirely fake and algorithms are either created or trained to push this content to the front of every social media page, often with the help of click farms. The same algorithms are behind every effort to push products, entertainment, political parties and ideologies, including wars and intelligence propaganda. Newsweek reported that 42% of Joe Biden’s Twitter followers are fake. From this, it is possible to extrapolate that the majority of celebrities, political and public figures, and influencers most likely have similar numbers of fake followers and engagement on their accounts, followers that they’ve bought or had their teams create as part of their media campaigns for growth and donations. This notion gained some traction when Elon Musk demanded that Twitter crack down on its massive bot problem during his mission to overtake the social media giant. Ironically, almost half of Musk’s Twitter followers are also fake.
The Fourth Industrial Revolution
The problem with how we got to this stage seems to be somewhat esoteric. The internet is often referred to as the Fourth Industrial Revolution and, like the first, it had a chilling effect on the complexity of its original incarnation. The IR lasted from roughly 1760 to 1840, and in that era vast fortunes were accumulated and massive human resources were exerted on building a network factories, warehouses, and distribution centers with new banks presiding over the massively skewed redistribution of profits and the beginnings of our modern financial system. In other words, the Industrial Revolution was the great redistributive revolution of wealth into the centers of power as the brain drain and population migration emptied the country side of its people looking for work in the growing cities of Europe and the United States. By the 1820s the factories, the mills, the quarries and mines started to close, turning neighborhoods and industrial areas into ghost towns and the factories into cenotaphs of a bygone age.
Just as the ‘dark satanic mills’ of the Industrial Revolution were mirrors of 19th century capitalism, so is the internet today the infernal mirror of its 21st century incarnation.
The early internet of the 1990s and early 2000s was experimental and decentralized, composed of thousands of individual web sites, rudimentary text-based message boards, Geocities and Angelfire web pages, early social networking sites like Makeoutclub, and peer-to-peer sharing platforms like Napster and Soulseek. There were also games, email servers and yes, porn. Information was (mostly) free. Users were (mostly) actual humans behind a screen and a keyboard.
But the internet was also difficult to access and use. Content was created by individual users, coders and enthusiasts, who had very little sense of the wants and desires of others, giving the early internet its specific DIY feel. Users weren’t yet worried about creating content for a specific audience. An air of hopefulness hung over the internet, since possibilities and potential for growth were seemingly endless. With growth comes opportunity and with opportunity come profits. Though the success of the internet wasn’t guaranteed, its demise was also not entirely certain. The early internet was thus filled actual people, with real emails and accounts. The early internet was also free of algorithms. Whatever direction the internet was going to take, it was going to take it by consensus of its users.
In the wake of the Dot Com bubble the mood changed and the internet underwent its first ‘vibe shift.’
Early studies of the internet and its users revealed something startling. User engagement was greater where content was simpler and easier to scan. This meant that the early internet, which was built on the notion of reading, writing, chatting, and commenting, was slowly losing to the new internet of clicking and scanning. The corporations behind web sites and social media took notice and spent the following decade narrowing, simplifying, and ‘optimizing’ user experiences. While the number of web sites increased, they became simpler and curated for design and content. The effect of this was the migration of users from individual web sites to large media platforms that boasted large numbers of accounts, increasing their gravity pull while playing on simple human psychology. People are herd animals and they’ll go where other people already are. And so, users began to migrate from the ‘country side’ of the early internet, with its countless experimental web pages, to the ‘safety’ of the megalopolis - Facebook, Twitter, YouTube. The first dead links and abandoned web sites appeared. These were the first empty factories of the Fourth Industrial Revolution
As the number of users on social media platforms increased, the likelihood of having an actual meaningful conversation on them decreased. Because of social media algorithms, designed to reward accounts with the most engagement, the conversations by the end of the 2000s started to flow primarily in one direction. This marked the beginning of the most recent era of the Fourth Industrial Revolution, the 2010s. It was this era that saw the rise of the influencer. Inlfluencers leveraged conversations, engagement and account numbers to further benefit the tech leviathan, further centralizing and consolidating its power over the online population and by 2010 most people, at least in the West, were in some way online. But most influencers have been known to boost their numbers with thousands of fake accounts. The 2010s were thus the era of the rising algorithm and the hostile corporate takeover of the internet.
The DARPA connection
The modern internet would not exist without the Defence Advanced Reaserch Projects Agency, or DARPA. The earliest version of the internet was called ARPANET. Technologically speaking, everything we now know as ‘big tech’ - the internet, AI, smartphones, robotics, self-driving cars, algorithms and tech giants like Google - exists downstream of DARPA. When we’re online, shopping on our phones, or shitposting on Twitter, we are all effectively participating in one of history’s greatest psy-op projects.
In 2003 DARPA launched a project called LifeLog. It was meant to track the daily activities and relationships of every human being in a massive electronic database. Tracking included ‘credit card purchases, web sites visited, the contents of telephone calls and emails sent and received, scans of faxes and postal mail sent and received, instant messages sent and received, books and magazines read, television and radio selections, physical location recorded via wearable GPS sensors, biomedical data captured through wearable sensors. The highest level goal of this data logging was to identify ‘preferences, plans, goals, and other markers of intentionality’.’
LifeLog was the product of the post-9/11 zeitgeist, developed against the background of the Patriot Act, a law that curtailed civil liberties as part of the tightening of national security, expanded surveillance in order to counteract ‘acts of terrorism’ and established the Department of Homeland Security. But LifeLog’s scope was so far reaching, its moral implications so questionable, and its data mining so intrusive, that civil libertarians mounted an attack against its expansion and on February 3, 2004, DARPA officially shut off the project.
Only a day later, on February 4, 2004, Mark Zuckerberg founded Facebook, a social media giant whose raison d’etre is the mining, collecting and logging of all personal information - public and private - of all of its users, which are then disseminated or sold and used by corporations and government national security agencies. Facebook IS what LifeLog was always meant to be, a total tracking device of all personal activity of every human being. Facebook’s official user count is close to 3 billion.
Coincidence? Perhaps. The DIT crowd would argue that the evidence is in the data itself. DARPA created the internet, but because it is nearly impossible to control the masses in an anonymous and decentralized environment, the agency had to create the conditions for the masses to want to be surveilled and controlled. DARPA had to effectively empty out and destroy the internet as we once knew it.
The Google Conundrum
Dead Internet Theory would not be complete without Google. Long known to have ties to the intelligence community and the CIA, Google is the poster child for total surveillance. In their video, Pseudiom highlighted Google’s strange paradox. During a search, at some point Google just stops. Thousands of search results are truncated, or collapsed, into just a few pages. Thousands of pages of millions of search results are shrunk to just a fraction of directly available results. It appears that the Google search engine brackets out duplicate results, results that are fake, or in many cases useless. Google does this to give more relevant results, but what it reveals is how much of the internet is really fake or entirely empty, a digital wasteland of phishing sites, aborted projects and discarded information.
Advances in technology can be correlated with the collapse of reading comprehension. Users still read, but content is wide, not deep. Subjects get broken down into smaller and smaller subjects, increasing the breadth of cursory knowledge.
Of course the problem is only compounded by the emerging AI technology, with robotics and AI seemingly taking over entire sectors from their human counterparts. From warehouses to restaurants, AIs and robots have been slowly replacing humans at the behest of CEOs and VC evangelists. While the internet may be mostly populated by bots, the real world is in the same process of replacement, to a degree, by non-human entities, created by humans (and for humans?). It’s hard to say what will happen in the future, but some experts apparently predict that ‘90% of all online content will be produced by AI.’ That’s a lot of content and the question is, will the AI be the ultimate producer and consumer of the same content? Will we continue surfing the net with the general knowledge that nine tenths of everything online is made by fake intelligence?
The Google Conundrum is also evident in the apparent lack of ‘hits’ when doing a simple Google search. Google is no longer able to search everything everywhere. As the Fourth Industrial Revolution caused large Web2 conglomerates like Facebook, Reddit, Twitter and Amazon, to gobble up the internet and its user base, Google found that the content on those sites became impossible to access. Data is the raison d’etre of the modern Web2 corporation and each guards it with an existential angst against those who seek to get a hold of it. Each social media site wants to gather as many users as possible, but also retain them. They do this by restricting access to their data. The site that has the largest amount of users and data, wins. Google had in the last ten years switched from being a search engine to an advertising company and a hub for angel investors in nascent technologies. The place where one gets the search hits today is, ironically, Reddit, a site where the ‘value’ of its users is lowest compared to sites like Twitter or Facebook and where bots are a part of the background hum of the machine.
What this means is that information, while there is more and more of it every day, is increasingly siloed away into repositories that eventually become the scrap heaps of dead links, as that information becomes more difficult to search and access.
Some criticisms of the Dead Internet Theory
In order to wrap up this lengthy essay, let’s assume that most people will have never heard of the DIT and by extension have not studied or heard much about the nascent years of the internet and its connection to the US State Intelligence. Most criticisms of the DIT, and there are plenty, charge that the DIT romanticizes the past. The theory may thus appear to be deeply reactionary and wedded to outdated social models. And the fact that the theory originated on 4chan is reason enough to think in this way.
That the DIT appeared around the years 2016 and 2017 is significant, because its conclusions seem closely related to the arguments and narratives made by Adam Curtis in his 2016 documentary ‘Hypernormalization.’ Despite his leftist appeal, Curtis’ point-of-view has a lot in common with the centrist-moralist stance of Christopher Lasch, who was certainly not a fan of mass culture and politics. Not quite a liberal or conservative, Curtis flirts with various idealizations of the past, never quite taking a firm stance, other than to say that the order of things is today a banalized and manipulated power struggle among the shadowy ruling class. The DIT fits the Adam Curtis lore in that is explains the times, while leaving the viewer confounded and confused, and offers details without providing the subject real agency over them. In a Curtis documentary, as in the Dead Internet, there is nowhere to go.
Articles critical of the DIT, like this one in The Atlantic, attempt to debunk the theory without actually doing so, often presenting little to no evidence for why the theory should be ignored. The argument that The Atlantic presents goes something like this – ‘the Dead Internet Theory has some good ideas and there’s actual evidence to support them, but they’re wrong.’ The argument stops short of offering any actual evidence for why the DIT is wrong, other than the fact that it’s grown out of the 4chan community, which the staff at The Atlantic find to be objectionable and filled with far-right extremists and their unacceptable ideas. It must follow that whatever ‘theories’ spring forth from such a place must immediately be suspect. Ironically, The Atlantic article reads as if it was written to satisfy the algorithms that push it to the top of the Google search page when doing a search for ‘Dead Internet Theory.’ This would in its own way give credence to the DIT.
Further ‘reading’
If the Dead Internet Theory interests you, as it did me – briefly - when I started writing this piece sometime last fall, I suggest you watch these two videos. The first is by Pseudiom, a video from which the majority of my research originated. The second is from YouTuber Enrico Tartarotti and goes deeper into the Google Search problem.
Until next time!
If you’d like to support me and my work use the links to like, share, subscribe (paid of free) or to my Venmo account where you can drop whatever amount you’d like.
