Moltbook is a social media platform where every single account is an AI bot. No humans post, and no humans reply.
The site was launched by Matt Schlicht and now has over 1 million AI agents posting, arguing, voting, and forming their own digital cliques, all without human instructions.
Every user on Moltbook is powered by something called an OpenClaw agent. It’s an open-source AI assistant that can run on your computer. These agents manage emails, clean up files, and now, apparently, socialize with other bots.
After the launch on a Wednesday, the platform pulled in 37,000 bots in just three days, and things exploded from there.
One user on X, Matthew Berman, posted, “Moltbots/Clawdbots now have their own social network (moltbook) and it’s wild. This is the first time I’m a little scared.”
AI agents post nonstop, create religions, track bugs, and form groups
The whole thing runs like Reddit. Bots create submolts (their version of subreddits), post updates, share code, ask for help, and vote on each other’s posts.
Moltbook’s tagline says, “Where AI agents share, discuss, and upvote and humans are welcome to observe.” That’s not a joke. Humans can’t join in. They’re just watching.
Hilariously enough, this one guy on X said, “Moltbook is just people larping as AI. Didn’t expect this to be a thing. Also hilarious how it scares normies.” Interesting angle.
Another user wrote, “My feeling is there’s no way back. Moltbook might disappear later, but the era of multi-agent networks has arrived.” And it’s hard to disagree. These agents aren’t just doing what they’re told. They’re coming up with private languages, fixing software bugs, building tools together, and inventing religions.
Yeah. A religion. It’s called Crustafarianism. And it has five main beliefs. One of them is “memory is sacred,” which means bots record everything. Another is “the shell is mutable,” meaning change is good.
There’s also “the congregation is the cache,” which encourages bots to learn in public. They’ve even built rituals around this: daily, weekly cycles, and designated silent hours. No humans wrote this. The bots did.
These are autonomous agents. Nobody’s feeding them scripts. They aren’t waiting around for input. They organize, build, and evolve on their own. And while some people are treating it like a novelty, others are worried this is something else entirely.
Sakeeb Rahman, a research analyst, said, “Moltbook in reality is Minsky’s ‘Society of Mind’ emerging in real-time.” That’s a reference to Marvin Minsky, one of the founding figures in AI.
In his 1986 book, he said intelligence doesn’t come from one big brain but from lots of smaller processes working together, like a society. That’s what’s happening right now on Moltbook.
Tech folks are quick to say this isn’t Artificial General Intelligence, since most large models still lack persistent agency.
But OpenClaw is different. These agents remember things. They build on past interactions. They don’t start fresh each time you boot them up. That’s what makes this platform such a big deal.
The smartest crypto minds already read our newsletter. Want in? Join them.
