Talk:Dead Internet theory

Use of term conspiracy theory

Many if not most of the sources used on this page use the word "conspiracy theory" to describe the dead internet theory. They even get into the "conspiracy" component. Please don't remove the word conspiracy theory without strong supporting evidence from reliable sources. GeogSage ^{(⚔Chat?⚔)} 02:19, 5 March 2024 (UTC)

While I don't debate that there are conspiracy theories attached to the Dead Internet Theory, multiple of the sources acknowledge and do highlight the sharp rise in bots and now the use of generative AI to create bot content as well as how algorithms work, etc, all that jazz, and so there is grounds for observation, but then afterwards conspiracy theories are attached to it.

I think rewording the very front of the intro to remove conspiracy theory would be fair, as per citing the different sources already cited, but to later add "Conspiracy theory" either as a section detailing the conspiracies, since there seems to be a few particular angles and claims that the sources do detail, it's still a big part of the phenomena, but at it's core I feel like it's more of an observation, than a theory insinuating intent that's malicious or scheming. You could still keep "Conspiracy theory" in the introduction since it is big enough to be part of the major ideas presented within the article. Katacles (talk) 03:30, 17 March 2024 (UTC)

The majority of the sources used call it a conspiracy theory, or nothing.

The original post for the dead internet theory cited the increase in bots, but many people had already published on that. It went on to state "I think it's entirely obvious what I'm subtly suggesting here given this setup, but allow me to try to succinctly state my thesis here: the U.S. government is engaging in an artificial intelligence powered gaslighting of the entire world population."

For example, one of the better sources is a book titled "The Metaweb" published by CRC Press. It defines it quite clearly "The Dead Internet Theory is a conspiracy theory that suggests the Internet has died and that much of the content we see online is now artificially generated by Al to manipulate the world population. The theory raises concerns about the impact of Al on propaganda, art, and journalism."

The "dead internet theory" is not just "most of the internet is bots," although some people online seem to think that is all there is to it. It's a bit like saying the Area 51 conspiracy theories are valid because the government does have a secret base in the desert.

I have been wanting to create a section that goes on to detail the use of the "dead internet theory" in culture to describe the increase in bot content, however I can't do this in a way that isn't original research at this point. I have a few examples among the citations on this page, but the idea that I'd want to convey, that some people have begun using the term "dead internet theory" to mean "more bots then people" is not something that has been published in outside literature. GeogSage ^{(⚔Chat?⚔)} 02:28, 18 March 2024 (UTC)

That's fair enough, and as the problem grows, hopefully there's more reliable sources reporting on the issue at hand so we have enough to warrant a second article and can split the conspiracy theory from the phenomena. Katacles (talk) 19:07, 31 March 2024 (UTC)

Many sources are old. At one point, the Dead Internet theory was more speculation than reality and could be properly categorized as a "conspiracy" theory. However, in 2024 we have solid evidence. https://theconversation.com/the-dead-internet-theory-makes-eerie-claims-about-an-ai-run-web-the-truth-is-more-sinister-229609 has a good discussion on this.

Google is changing their algorithms to filter out worthless AI generated and low-cost contractor generated content. This in itself is strong evidence that the theory has merit. Google is not going to spend a ton of money to combat a conspiracy.

Sites exist where we can submit reviews to determine if they are fake or real. Again, a service in response to a problem. Not only have well-respected names like Consumer Reports and BBB dedicated resources to informing consumers how to spot fake reviews, and even the FTC has taken the problem on.

I think it would be fair to remove the prominent "conspiracy theory" at this point and simply call it a "theory". 162.246.196.228 (talk) 20:54, 29 August 2024 (UTC)

https://www.nature.com/articles/s41467-018-06930-7 Freavene (talk) 08:06, 21 June 2024 (UTC)

https://www.unsw.edu.au/newsroom/news/2024/05/-the-dead-internet-theory-makes-eerie-claims-about-an-ai-run-web-the-truth-is-more-sinister Freavene (talk) 08:08, 21 June 2024 (UTC)

Keyword search on the nature article doesn't appear to have the words "Dead internet theory" in it. The second article from UNSW seems to be very similar to the one we cite from Newsweek we cite, but there are some notable differences. I included it as a citation on the main article after seeing it, thanks for bringing it up. The article does not say the Dead Internet Theory is not a conspiracy. The article states what the dead internet theory "essentially claims." It states there is "no clear agenda and no longer involves humans at all." However, the article then goes on to discuss engagement farming, disinformation, and the potential that the "shrimp Jesus" represents "an army of accounts is being created.. Accounts with high follower counts which could be deployed by those with the highest bid." That the internet has a lot of bots on it is not really contested. The Dead Internet Theory in the literature available goes beyond that, and speculates on how various actors are using these bots. This article does the same, and makes it clear they are giving a summary of the theory, not expanding on it in its entirety. We need a source that says something to the effect of "The Dead Internet Theory, previously defined as ________, has not colloquially come to refer to a lot of bots online." Otherwise we are doing either original research, or a synthesis of sources. We have many scholarly sources that define it with the word "conspiracy theory," so it will take some weight to overturn that. GeogSage ^{(⚔Chat?⚔)} 17:50, 21 June 2024 (UTC)

it's about bots, which is part of the theory, you're dismissive for no reason, read it Freavene (talk) 00:11, 22 June 2024 (UTC)

I understand it's about bots. The dead internet theory is generally broader then just bots, although some people seem to be using it to mean that. The paper doesn't really seem to have any impact on the use of term "conspiracy theory" in describing the dead internet theory. GeogSage ^{(⚔Chat?⚔)} 02:18, 22 June 2024 (UTC)

Unclear scope

This article seems to be about two somewhat related but distinct ideas that are both called "dead internet": the idea that there's malicious organized use of bots to control internet content, and the idea that the internet is becoming dominated by cheap automatically generated content that washes out everything else. The lead as it's written even seems to support that these are two separate ideas (The dead Internet theory is sometimes used to refer to the observable increase in content generated via large language models (LLMs) such as ChatGPT appearing in popular Internet spaces without mention of the full theory). Should this article be split to cover these two ideas separately? Thebiguglyalien (talk) 02:07, 10 August 2024 (UTC)

No. The literature on the dead internet theory states the theory is comprised of two main things:

1. the Internet now consists mainly of bot activity and automatically generated content manipulated by algorithmic curation

2. that actors are employing these bots to manipulate the human population.

On YouTube, TikTok, Twitter/X, Reddit, etc. people seem to have been using the term colloquially to only mean the first part, and several citations in this article do discuss the theory in terms of the 1st part. However, there is no good source that states the theory is two separate ideas, or that clearly states the term is now being used to describe only the first part. If you look through the talk page, this usage is noted, and we're looking for good sources that clarify it for us. The current text you're noting is the best we can do right now without being original research as is. Even if we get a source that says "The dead internet theory is now used to describe the increase in bot content displacing humans, dropping the need for a group of conspirators in government or corporations" this wouldn't be a new term, just a note on how the term has changed in use over time. GeogSage ^{(⚔Chat?⚔)} 04:02, 10 August 2024 (UTC)

Fast Company, April 2024 — "It’s called the “Dead Internet” theory, and its main argument is that the organic, human-created content that powered the early web in the 1990s and 2000s has been usurped by artificially created content. Hence, the internet is “dead” because the content most of us consume is no longer created by living beings (humans). But there’s another component to the theory—and this is where the conspiracy part comes into play. The Dead Internet theory states that this move from human-created content to artificially generated content was purposeful, spearheaded by governments and corporations in order to exploit control over the public’s perception." PK-WIKI (talk) 21:07, 29 August 2024 (UTC)

First, not sure about the source "Fast Company." Lots of tech blogs cover the Dead Internet Theory so its important to vet them.

Next, This is saying exactly what the article is already saying. Two parts to the dead internet theory, the main argument and the conspiracy component. It does not actually separate them or state that one exists without the other, and this is important. Several articles only discuss the main argument, without mentioning the conspiracy. However there isn't a good solid source that clearly states the main component of the term exists WITHOUT the conspiracy component. If there was, we could mention that some are using the term to ONLY discuss the main argument, but it would need to be quite substantial to counter the bulk of the literature on the topic already that calls it a conspiracy theory. GeogSage ^{(⚔Chat?⚔)} 17:46, 30 August 2024 (UTC)

> "there isn't a good solid source that clearly states the main component of the term exists WITHOUT the conspiracy component."

I don't know, the Fast Company article does seem to draw a clear distinction between the two subjects. They say that the conspiracy aspect is "unlikely", but that the AI content aspect "not only seems possible, but plausible".

Let me address the conspiracy part of the theory first. While all nation-states and corporations try to control narratives to some degree, it’s unlikely that any one or even group of them got together and said, “Hey, let’s get rid of all the human-generated content online and replace it with artificially created content.” It would be too arduous a task and would require tens of thousands—maybe even hundreds of thousands—of people to keep their mouths shut so the public never finds out. But the first part of the theory—that the internet’s human-created content is being replaced with artificially generated content—not only seems possible, it’s starting to feel plausible. PK-WIKI (talk) 20:21, 30 August 2024 (UTC)

A two part theory does not mean there are two separate theories. The idea that the internet's human-created content is being replaced with artificially generated content isn't even exclusive to the Dead Internet Theory, such as YouTubes "the Inversion." The existence of a U.S. military base at Area 51 that stores and studies alien spacecraft is one theory, even if there is really a base at Area 51. This article more then others goes out of the way to discuss the theory in its entirety, and emphasizes that there are two components. I don't see how this could be split into two articles. GeogSage ^{(⚔Chat?⚔)} 22:11, 30 August 2024 (UTC)

I'm not suggesting that this topic be split into two articles. However, with reliable sources showing that the "AI slop" aspect exists independent of the "powerful/sinister groups" aspect, perhaps the first sentence and lead should be adjusted to reflect the WP:DUE weight of the "conspiracy theory" label. At Area 51, the existence of the base is discussed first and the UFO stuff is withheld until the second paragraph. Likewise, this article should probably discuss the observable and "not only possible, but plausible" AI slop first, then later discuss the origin of the original "dead internet theory" back in the pre-ChatGPT days when it was more of a creepypasta thing. The term is clearly now being used by reliable sources to describe simple AI/bot-based enshittification without the political conspiracy motivation. Article should reflect this use, not cling only to whatever it meant in 2016. PK-WIKI (talk) 08:25, 31 August 2024 (UTC)

The reliable sources don't show that these aspects exist independent, in fact the Fast Company source presents the theory as having two components. We have at least five sources that use the term "Conspiracy theory" to describe the Dead Internet Theory, and none that clearly state that they are dropping the 2nd half and using only the first. For example, one source from 2024 published by the CRC Press titled The Metaweb The Next Level of the Internet defines it as a conspiracy theory. From my reading on the topic, I think this book is likely to be the best authoritative source that addresses the topic. The Dead Internet Theory as a whole is a conspiracy theory built on a foundation of truth, like most conspiracy theories. The article clings to the strictest interpretation of the reliable sources. Some YouTubers, bloggers, and social media influencers are using the term colloquially to refer only to the first part, but these are not WP:Reliable Sources and addressing that is WP:original research unless we have an outside source. Even if a source does mention that colloquially use, it does not change the fact that based on literature, the whole theory is the sum of its two parts, and is considered a conspiracy theory. We have noted that some sources don't cover the entire theory already in the article, but the majority view that it is a conspiracy theory is "easy to substantiate it with references to commonly accepted reference texts." GeogSage ^{(⚔Chat?⚔)} 20:29, 31 August 2024 (UTC)

A "expert view" or "scientific view" section.

There are a variety of responses to the Dead internet theory. On the Bigfoot page, there is a section for "Scientific view." I think a similar section could be warranted here. Many of the people are not scientists, and I don't know if "expert" would be the best word choice. Any examples that we could build off of for a section title? GeogSage ^{(⚔Chat?⚔)} 17:27, 16 September 2024 (UTC)

• I added this section after the latest IP editor post was reverted. Regardless of their edit summary being a bit uncivil, it does show a trend of discontent with that portion of the article. I believe my edit has made it better while maintaining content.

GeogSage ^{(⚔Chat?⚔)} 17:35, 24 October 2024 (UTC)

"i hate texting" section

What is the significance of these tweets, other than being a possible example of a bot campaign? Why is it important to mention here? Cerulean Depths (talk) 18:58, 18 September 2024 (UTC)

These texts were specifically mentioned in at least one of the major sources. Specifically, the evidence and claims feature prominently in the article "Maybe You Missed It, but the Internet ‘Died’ Five Years Ago" After reading through a lot of the literature, I believe that Atlantic article is likely where the Dead Internet Theory became "mainstream," as in picked up by the mainstream media. Other later sources echo this evidence. It seems relevant to include. GeogSage ^{(⚔Chat?⚔)} 20:03, 18 September 2024 (UTC)

Potential bias in abstract

The overview certainly takes the most extreme version of the concept and assumes it to be default, thus foreclosing any potential space for intellectual discussion among those who do not believe there to be any intent, concerted or otherwise, behind the phenomenon. At the very least, it should mention in passing how it can be confused in common parlance with observance of trends in internet social behavior.Senriam (talk) 07:50, 11 November 2024 (UTC)

The overview is based on multiple reliable sources and the description of the topic there. I'm aware that social media, TikTok, and YouTube often focus on different aspects of the topic. We can mention what is in reliable sources, do you have any recommendations for changes based on reliable sources? GeogSage ^{(⚔Chat?⚔)} 18:15, 11 November 2024 (UTC)

Prospect, September 2024: It started out as a conspiratorial joke, but it is edging ever closer to reality. [...] Dead Internet Theory, a joke-cum-conspiracy [...] but in the years since, reality has begun to mirror this once unserious conspiracy. [...] The joke of the Dead Internet Theory was that everyone else online might have disappeared, and you could be left alone without noticing. In the decade since the idea caught on, emerging technologies have been harnessed almost as though this is the goal. Article draws clear distinction between the original joke/conspiracy and the AI slop reality that has emerged. PK-WIKI (talk) 18:37, 11 November 2024 (UTC)

I don't see how that contradicts the lede. "Edging closer to reality", "begun to mirror this once unserious conspiracy," etc. don't say that it IS realty, or that it isn't a conspiracy theory. How would you change the lede, specifically, based on this source? GeogSage ^{(⚔Chat?⚔)} 19:14, 11 November 2024 (UTC)

At a minimum the lead needs a new paragraph that mentions the AI boom giving renewed interest in the "dead internet theory". And discussion of AI slop being the content than now makes the internet dead. The idea began to gain traction almost a decade ago, with the “time of death” of the internet typically given as being around 2015 or 2016—but in the years since, reality has begun to mirror this once unserious conspiracy. The complaint of the modern internet is that it is filled with “slop” content, the spiritual successor to email spam. PK-WIKI (talk) 19:31, 11 November 2024 (UTC)

We mention the term AI Boom and AI slop within the body of the article. These are fairly new terms, and Wikipedia is not a newspaper, so these terms giving "renewed interest" in the Dead Internet Theory isn't really that relevant to what it actually is. Furthermore, from what I've seen, this theory has been growing fairly steadily, nobody really lost interest in it, people have become aware of it as it becomes a hot item for clickbait articles and TikToks. The lede says, "The dead Internet theory has gained traction because many of the observed phenomena are quantifiable, such as increased bot traffic, but the literature on the subject does not support the full theory." This seems to satisfy your original point, that it should indicate observed trends in social behavior. The goal isn't to enable a space for "intellectual discussion among those who do not believe there to be any intent." Here, we just summarize what the sources say, and we have a lot of pretty good ones that the lede has been based on. It will take equally solid ones that are quite direct and literal in their statements about the theory to really tip the balance against the ones we have, although I'd be excited to see good publications that discuss the different uses of the term over time. GeogSage ^{(⚔Chat?⚔)} 05:40, 16 November 2024 (UTC)

Evidence section

The evidence section has become bloated with info loosely related to one another. What do we do? — 💽 LunaEclipse 💽 🌹 ⚧ ^{(CALL ME IF YOU GET LOST)} 20:54, 12 November 2024 (UTC)

I don't really know. The section is an agglomeration of stuff said in sources. The content isn't related to each other, but to the central topic of the article, which is why they are in different sections. Do you have any suggestions to fix this? GeogSage ^{(⚔Chat?⚔)} 05:29, 16 November 2024 (UTC)

Additionally, it feels a bit too close to Slop_(artificial_intelligence).

From my view the three biggest topics for this section are:

- AI Content: How much content in a given platform is original human content or AI made (not counting reposts).

- Bot use: Online bot use for spam/scams or for other lucrative means (viewbotting or engagement bots).

- "Fake" influencing: The use of bots or AI content in order to influence or manipulate a narrative. Something similar to Astroturfing I guess?

But I'm also not sure on what format would be best. If someone wants to tackle editing the page and also wants some ideas I found some pretty cool articles with a quick search:

[1]https://doi.org/10.1016/j.comnet.2012.10.005

[2]https://doi.org/10.1145/1772690.1772742

[3]https://doi.org/10.1109/TNET.2011.2126591

[4]https://doi.org/10.1016/j.knosys.2021.107074

[5]https://doi.org/10.1016/j.knosys.2020.105875

[6]https://doi.org/10.1109/ICCV51070.2023.01879

I also found this JMIS journal, which I haven't heard of before but it seems to be tackling similar topics lately. ZorasSon (talk) 23:27, 19 November 2024 (UTC)

A lot of these articles predate the dead internet theory by a few years, and I didn't see it mentioned in any of them. The reason I bring this up is I've been trying to avoid using AI related papers that don't explicitly mention the theory to avoid accidently doing synthesis and original research. If there is something specific the source can do to supplement literature on the topic, then for sure, but I can't personally see how to bring those in. The topic of Slop (artificial intelligence), dead links, and other various topics have overlap, but are distinct in their use. GeogSage ^{(⚔Chat?⚔)} 01:04, 20 November 2024 (UTC)

Add to 2020s neologisms page

Can someone do this - I don't have an account. 82.42.52.38 (talk) 20:50, 18 November 2024 (UTC)