kb/data/en.wikipedia.org/wiki/Dead_Internet_theory-1.md

6.0 KiB

title chunk source category tags date_saved instance
Dead Internet theory 2/4 https://en.wikipedia.org/wiki/Dead_Internet_theory reference science, encyclopedia 2026-05-05T07:10:21.240001+00:00 kb-cron

=== "Leaner" dead Internet theory === In a 2026 article, Hal Berghel discussed what he called a "Leaner" dead Internet Theory, "stripped of paranoia, prejudice, politics and polemic." In this discussion, Berghel points to the 2025 publication in AI & Society by Yoshija Walter, and lists algorithmic generated content, Generative AI byproducts, the difficulty for some people to distingusish between this and human generated content, and the resulting mistrust and misinformation as the core of the dead Internet theory. Berghel laments that conspiracy theorists take these phenomena and to make implausible claims, while arguing that core criticisms should not be dismissed.

== Expert view == Caroline Busta, founder of the media platform New Models, was quoted in a 2021 article in The Atlantic calling much of the dead Internet theory a "paranoid fantasy", even if there are legitimate criticisms involving bot traffic and the integrity of the Internet, but she said she does agree with the "overarching idea". A 2021 Ouest-France article, which heavily referenced the 2021 article in The Atlantic, went on to compare the dead Internet theory to flat Earth and COVID-19 conspiracy theories. The article stated that even though bots do produce online content, the dead Internet theory is still not realistic. In 2023 in The New Atlantis, Robert Mariani called the theory a mix between a genuine conspiracy theory and a creepypasta. A 2023 book published by CRC Press discussed the dead Internet theory, specifically mentioning Google censoring the web. The book included an entry for the term in its glossary defining it as:

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 AI to manipulate the world population. The theory raises concerns about the impact of AI on propaganda, art, and journalism. A 2024 IFLScience article stated:

Like all good conspiracy theories, the Dead Internet Theory takes a kernel of truth or agreed sentiment (that the internet is getting worse, and that bot activity is increasing) and twists it into something it isn't. In 2024, "dead Internet theory" was 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. In a 2024 opinion column in AI & Society's "Curmudgeon Corner", Yoshija Walter stated that the once speculative theory is now observable with the introduction of AI generated content. In a 2025 article by Thomas Sommerer, this portion of the dead Internet theory is explored, with Sommerer citing Walter and calling the displacement of human generated content with artificial content "an inevitable event". Sommerer states the dead Internet theory is not scientific in nature, but reflects the public perception of the Internet. Another article in the Journal of Cancer Education discussed the impact of the perception of the dead Internet theory in online cancer support forums, specifically focusing on the psychological impact on patients who find that support is coming from an LLM and not a genuine human. The article cited both Walter and the CRC Press book when defining the dead internet theory, but did not mention the conspiracy aspect. The article also discussed the possible problems in training data for LLMs that could emerge from using AI-generated content to train the LLMs. In a 2025 paper, Roland Leikauf described the dead Internet theory as "pseudoscientific" while questioning if new AI tools would justify our fear that the theory might become reality. Leikauf cites Walter's 2024 publication for his definition of the dead Internet theory. In a chapter of the 2025 book Market-Oriented Disinformation Research, it states:

What makes the Dead Internet a nameworthy conspiracy is that even though it is rooted in selective truths that are exaggerated or even taken to their logical extremes, it also draws attention to a legitimate problem. In a 2025 interview with Time, linguist Adam Aleksic stated that the dead Internet theory "used to be a lunatic fringe conspiracy theory, but it's looking a lot more real". In a 2026 paper published in Computer (magazine), Hal Berghel discusses a "leaner" version of the theory, without the conspiratorial elements, focusing on the core claims. In this paper he states:

We must admit that some of the core principles of the DIT are convergent with our technical and historical experience. Unfortunately, the conspiracy theorists augment these very plausible observations with their own mix of biases and agendas that lead to implausibility and absurdity, which in turn leads to rejection. But it is a mistake of the first order to dismiss the core criticisms unequivocally.

== Criticism == The dead Internet theory has been cited by some as a conspiracy theory lacking credible evidence. While research has shown that automated bot accounts makeup a significant portion of internet traffic, experts emphasize that this provides no proof of a coordinated effort to replace human activity online. Critiques argue that the theory is based largely on unverified observation, such as perceived declines in content quality or repetitive online interactions, rather than verifiable data. Researchers also note that many forms of bot activity such as spam filtering, Denial-of-service attacks, and web scraping serve specific functions, sometimes maliciously, but do not indicate in any way that human participation on the internet is being replaced. Researchers have also compared the dead Internet theory to other internet related conspiracy theories and note that these theories often stretch current trends, like artificial intelligence and increased bot activity, into speculative claims without using concrete evidence.

== Evidence ==