The 'Dead Internet Theory' posits that the internet has been largely 'dead' since around 2016 (at the earliest). This means that the largest share of online content and interaction no longer comes from humans, but from bots and AI. Proponents of the theory say that a large part of what we read, like, or share is generated or pushed by algorithms, government programs, or marketing machines. The theory originated on forums like 4chan and Reddit and has since become a fascinating topic of discussion.
Bot traffic and AI content: facts vs. speculation
Although the conspiracy aspect is high, the theory contains a kernel of truth. Recent studies show that bots account for more than 50% of all web traffic. According to data from Imperva, nearly 50% of all traffic in 2024 came from non-human sources, a significant portion of which involves 'bad bots' aimed at scraping, credential stuffing, and spam. But that is one source - some claim that humans have been the smallest group on the internet since 2016.
With the rise of generative AI like ChatGPT, Claude, Gemini, and SEO bots, it is possible to produce vast amounts of content at a rapid pace, often without clear attribution or quality control. This abundance, also known as AI slop, reinforces the feeling that the internet is no longer fed by humans, but by machines speaking in algorithmically optimized sentences.
Risks of a dead internet
Whether you believe in the theory or not, the phenomenon of an increase in (AI) bots is already causing a number of problems. Misinformation and content pollution are increasingly flooding facts, especially now that it is becoming harder to recognize what is and isn't AI. This also leads to crumbling trust. Whether it's your crazy aunt or a journalist, distrust in people is clearly growing.
But continuing to verify and investigate becomes a problem as well. People suffer from search fatigue: endless quests for the right information, without really knowing if you've found something real. It is becoming more and more difficult to distinguish valuable or original information from synthetic bulk. AI and the Dead Internet Theory itself both play a role in this; one produces far too much content, the other assumes that nothing is real.
The role of AI in the Dead Internet Theory
Generative AI has made creating texts, images, and videos easier than ever. With the help of LLMs, blogs, product descriptions, (fake) reviews, and FAQs are pumped out like oil from the ground. That content is then optimized based on search engines, further burying authenticity. AI thus significantly feeds the Dead Internet Theory.
Additionally, there is the phenomenon of synthetic engagement: bots that simulate likes, comments, and shares to influence algorithms. AI and bots set to generate engagement do just that; they do not pay attention to who is participating, what is being said, and so on.
Dead Internet: conspiracy theory or warning?
The Dead Internet Theory serves as a mirror for the state of the internet today: abundant, filtered, and increasingly non-human. Whether the majority of the internet is actually dead, with corners where bots only talk to each other, is of course the question. We can at least conclude that the online world is a world unto itself. AI helps the Dead Internet Theory gain more traction.
