In recent weeks, Reddit co-founder Alexis Ohanian and OpenAI co-founder Sam Altman have issued social media warnings about the “Dead Internet Theory,” which holds that bots, rather than humans, control the internet. However, experts who previously dismissed the theory as a conspiracy theory are now cautioning that, given the development of artificial intelligence, it may actually be true.
“Is this the time where, yet again, I remind y’all about Dead Internet Theory?” Ohanian stated on Sunday in a post on X, referencing another post on how many Reddit users were deceived by a weekly series of postings about an overweight cat on a weight loss journey dubbed “Pound Cake,” only to discover the cat was actually AI created.
Ohanian’s stance on the Dead Internet Theory has grown in recent months. In a June post on X, he cautioned that “it’s not *if* most of what we see online will be AI-generated.” The truth is that it already is.
Speaking at a panel in the Wall Street Journal in June, Ohanian admitted that he has “long subscribed to the dead internet theory,” which was considered a conspiracy ten years ago but is now “a very real thing” due to the rise of social media bots and the use of artificial intelligence (AI) by people to produce and spread content.
Altman has also issued a warning on the Dead Internet theory. In a September article on X, he stated that while he never truly believed in the idea, it appears that a significant number of LLM-run Twitter accounts are suddenly active.
51% is a big number. According to cybersecurity firm Imperva, it is the proportion of internet traffic produced by bots rather than people in 2024. According to the business, the growth of AI and Large Language Models (LLMs) has made it easier to create and scale bots for malicious reasons, which is why the sum was the first time in ten years that bot activity exceeded human activity.
The Dead Internet Theory: What Do Experts Say?
Some people rejected the Dead Internet theory as having little value prior to the widespread use of large language models, but in recent years, certain experts have said that the theory is increasingly likely to hold merit as artificial intelligence has grown. In 2021, Caroline Busta, the founder of New Models, a German company that investigates how technology affects culture, told The Atlantic—in a story titled “The ‘Dead Internet Theory’ Is Wrong But Feels True” that while she agreed with the “overarching idea” that the internet feels more “empty” now than it did ten years ago, some aspects of the theory are “paranoid fantasy.” Last year, a group of experts from Swiss institutions examined the Dead Internet Theory and concluded that, while it “used to be rather speculative” ten years ago, “it can now be observed first-hand with the wake of generative AI.”
According to the researchers, the proliferation of deepfake films has made it more difficult for people to distinguish between material produced by AI and content created by humans. The researchers also mentioned how generative AI is being utilized by retail businesses to sell goods, adding that social media is now more about consuming material and being addicted by specifically designed dopamine hits in our brains than it is about fostering human connections. AI bots can produce artificially generated images that are boosted and reposted by other AI-powered accounts, creating a vicious cycle of artificial engagement that “no longer involves humans at all,” according to an article written by Jake Renzella, a computer science lecturer at the University of New South Wales, and Vlada Rozova, a machine learning fellow at the University of Melbourne.
They referenced odd viral content such as the 2024 Facebook sensation “Shrimp Jesus,” an AI-generated picture. According to Alex Hern, the technology editor for The Guardian, Elon Musk’s X is also to blame for making the phenomenon more widespread. He claims that Musk made it profitable to purchase a blue checkmark, attach it to a large language model, and then set it to reacting wildly to viral content.
A surprising Fact
In order to investigate whether AI-generated bots might effectively influence users’ opinions on controversial subjects, a group of academics from the University of Zurich placed the bots on the well-known subreddit r/changemyview. Reddit retaliated by threatening legal action. The researchers never had their study published.
Key Background
In the late 2010s and early 2020s, the “dead internet theory” gained widespread traction on websites and forums such as Reddit. A member dubbed IlluminatiPirate posted a detailed idea on the internet’s death on an online forum. The post was highlighted by The Atlantic and had over 362,000 views. According to the user, the majority of the “supposedly human-produced content” on the internet is created by artificial intelligence, and the internet is “empty and devoid of people.” According to The Atlantic, the Dead Internet Theory also attracted interest in other online communities, such as the subreddit for well-known podcaster Joe Rogan, subreddits about paranormal phenomena, and a forum for followers of the well-known YouTube channel Linus Tech Tips, which focuses on technology.
