Yet another conspiracy theory-turned-academic curiosity^[1] has blossomed online, this one about the internet itself—and of course, AI.

Dead Internet Theory posits, that so prevalent and prolific are the automated posting systems, that any random sampling of internet content at the time of this writing, say mid-2026, would yield a roughly fifty-fifty chance that the content was generated by a bot.

To be clear, a lot of content is generated by bots. I proudly proclaim on this site that I write all the content I post here… but then I have systems and tools built in the back end that feed social media queuing platforms to help me promote those same words. Technically speaking, if you came to this post via social media, likely you clicked on a link posted by my bot.

Bots are not necessarily AI, though. Those posting systems of mine are merely complicated algorithms that do scheduling and pinging of updates (because I am terrible at remembering to do that on my own.) Little more than a calendar hooked up between my blog and my feed. Heck, I even built one.

But the difference between my innocence-pleading automated systems on one end of the spectrum and the slop-spewing generative content farms on the other is only one of semantics.

It’s all the same fever dream rainbow, after all.

The thrust of the Dead Internet Theory conspiracy is empowered by the vibes so many of us feel when we spend any amount of time online, though, that eerie feeling nudging our hearts that at some point we have stopped interacting with other people and might just be spending most of our day chatting with language models, reading posts created via prompts, or discovering newfound abilities to detect if that video voice-over is a human or a machine. 

There’s so much inhuman content online, it’s now a coin toss in knowing that what we encounter might just be some haunted undiscovered country of algorithmic automation that we are the first humans to lay eyes upon. Conspiracy indeed.