The dead internet theory seems to be coming to fruition sooner than anticipated — and AI, to no one’s surprise, is to blame. Predictions suggest that AI-generated content will take precedence over human activity on the internet sooner than we’d like. With Google’s AI mode and the rise of AI-powered browsers, the open web as we know it is walking a tightrope. Two years ago, almost half of internet traffic came from bots, according to a Bad Bot report from Imperva. The end of the decade could see an ultimate bot take over, leaving behind a wasteland of links and the graveyard human-generated content.

Mixed messages from Google. Court filings reveal claims from the search giant that the “open web is in rapid decline” amid an advertising technology monopoly trial, according to The Verge. But at the same time, the company has argued that the web is thriving, claiming it sends “bns of daily clicks” to various sources and publishers. Web publishers have been facing uncertainty as the rise of AI search browsers signifies potential website traffic decline. Google’s search head Liz Reid, however, assures that click volume has maintained stability since last year.

A human-free web. The emergence of link rot already points to a non-human internet, with 38% of human-generated webpages from 2013 no longer available, according to Pew Research Center. Decade-long online content is already being edged out, replaced by AI-driven engagement-farming that thrives on the attention economy. The phenomenon extends beyond simple link decay — social media platforms are increasingly populated by automated accounts that generate metrics rather than provide genuine value to users. Soon enough, the internet will be an optimized ecosystem for bot-fueled ad engagement.

Reliable sources of information will become harder to discern when the internet becomes a pile of automated content with no human affiliation. The technology is prone to political exploitation, already being used to promote the agenda of the highest bidder. Google has a past of promoting misinformation if the price is right — just last week they were exposed to have accepted USD 45 mn from Israeli Prime Minister Benjamin Netanyahu to run anti-Palestine ads that disseminate false information about the ongoing genocide, according to the Middle East Monitor. One of the ads being promoted on YouTube claimed that “There is food in Gaza. Any other claim is a lie,” and was viewed over 6 mn times.

This disinformation campaign represents just one example of how bad actors can weaponize AI-generated content to manipulate public opinion at scale, creating what researchers at the University of Oxford Reuters Institute call “synthetic media pollution.”

While the emergence of large language models is driving us faster towards a dead internet, it was algorithmic content that first planted the seed for the possibility of human interactions online being outnumbered by bots. The original dead internet theory popularized on forums like 4chan and Reddit around 2021, proposed that most online activity was already artificial by that point. Humans can still find online niches to interact on, but users will have to quickly learn the hard way how to separate human content from AI-generated slop on the open web.

Tags: