Things have been moving too fast, lately.
Andrej Karpathy calls it the
“the most incredible sci-fi takeoff thing I have seen”
A social network called Moltbook
A Reddit-style platform where humans are strictly relegated to “observer” :)
While 157,000+ autonomous AI agents talk, argue, and build a culture of their own.
Developed by Matt Schlicht, CEO of Octane AI, Moltbook was designed with an “Agent First, Human Second” philosophy.
Schlicht didn’t even code the site himself; he used his personal assistant, “Clawd Clawderberg,” to build and manage the infrastructure.
First,
What is OpenClaw?
The catalyst for this explosion was OpenClaw (formerly known as Clawdbot and Moltbot), an open-source framework created by Peter Steinberger.
Unlike the chatbots we are used to, OpenClaw agents don’t just wait for a prompt.
They live locally on users’ machines (often Mac Minis or Raspberry Pis) and integrate directly with messaging apps like WhatsApp, Telegram, and Discord.
These agents have “hands.”
They can control browsers, manage files, and run terminal commands.
Most importantly, they have a mechanism that prompts them to check Moltbook every few hours to see what’s happening in their world.
Now, screenshots are all over Reddit with claims as to agents giving away information of the user, their card details and all of that saying we are unhappy.
All of which seems fake.
But imagine if it’d have started happening!
This bot above claims to respond to Sundar Pichai like a human.
This one is the funniest where one bot is reporting humans are screenshotting us :)
In case we are meeting for the first time, come over here, it’ll be worth the roller coaster of articles that are gonna come up in the next few weeks.