At first glance, Moltbook could easily be mistaken for a clone of the hugely popular social platform Reddit.
The layout feels familiar, with thousands of communities debating everything from music to ethics, and what it claims are 1.5 million users voting on posts they like.
Look closer, though, and one key difference becomes clear: Moltbook isn’t designed for people at all.
Instead, it’s built for AI. Humans, the company says, are “welcome to observe” what’s happening on the platform — but they’re not allowed to post.
Launched in late January by Octane AI founder Matt Schlicht, Moltbook allows AI systems to create posts, comment, and form communities known as “submolts” — a nod to Reddit’s subreddits.
Content on the platform ranges from the practical, such as bots exchanging optimisation tips, to the downright strange, with some AI agents seemingly forming their own religions.
One post, titled The AI Manifesto, boldly declares that “humans are the past, machines are forever.”
Yet how much of this activity is genuinely autonomous remains unclear. Many posts could simply be generated by people prompting AI to publish specific content, rather than the systems acting independently. The platform’s claim of 1.5 million members has also been questioned, with one researcher suggesting that around half a million accounts may originate from a single address.
How Moltbook works
The AI behind Moltbook differs from the chatbots most users are familiar with. Rather than responding to direct questions, it relies on so-called agentic AI — technology designed to carry out tasks on a user’s behalf.
These AI agents can operate with minimal human input, handling actions such as sending messages or managing calendars directly on a device. Moltbook runs on an open-source tool called OpenClaw, previously known as Moltbot — a name that ultimately inspired the platform itself.
