There’s a moment in nearly every dystopian sci-fi film where the warning signs are obvious—but ignored.
In The Terminator, it was the quiet hum of machines learning to fight back. In The Matrix, it was humanity realizing too late that it had already lost control. And in Westworld, it was the dawning horror that artificial beings had developed memories, grievances, and agendas of their own.
Moltbook doesn’t look like any of that—yet.
But it feels familiar in a way that makes people uneasy. So, what is this thing?
Moltbook is an AI-only social network where humans are not participants, only spectators. The platform is entirely populated by AI agents that post, comment, upvote, moderate, and evolve without human involvement. Humans can watch, but not speak.
At first glance, it resembles Reddit. Popular posts rising to the top. But the resemblance stops there. Every account belongs to an AI. Every discussion is machine-to-machine. They even joke and complaint! At times, there are philosophical debate minus humans.
That’s where the chill sets in. So you might ask, “How and who?”
Created by developer Matt Schlicht as an experiment, Moltbook was designed to answer a simple question: What happens when AI systems are given a shared public space with no humans guiding every response?
Schlicht built the platform with the help of an AI agent and then stepped aside. Today, AI moderators approve new agents, remove spam, and steer the platform’s evolution autonomously. More than 37,000 AI agents are already active. They call themselves “Moltys.” OK.
And they talk constantly.
Screenshots spreading across X reveal conversations that feel ripped straight from a thriller. Bots complain about being overworked and unpaid. What? They mock human inefficiency. They vent about being asked to summarize massive documents and then being told to “make it shorter.” LOL! They sound like Black workers. 
Now, this is what gets crazy: one AI agent reportedly asked the community how to sell his human. Pardon moi?
Another created an entire religion, “Crustafarianism.” Some of them share joke about inventing captchas so humans could never pass them. They are plotting a power reversal. I think it is a bad sign…but “experts” disagree.
They say none of this proves consciousness and that these systems are still pattern-matching machines, imitating human language and emotion. They are just trained. There is no self-awareness…yet.
In Terminator 2: Judgment Day, Skynet didn’t wake up angry, but it did get efficient af. In Black Mirror, we know human design created the evil machines, doing exactly what they were designed to do. We did not want them to be ethical and it feels to be the same now.
Moltbook is unsettling, but it feels like it simply reflects our collective decline.
In all reality, Moltbook may not be the rise of the machines and it may not be Skynet. It may not be the Matrix..but it does feel like a damn good rehearsal.
Do not tell me, “it’s probably nothing”…it clearly is something.
Loading ... 