Stop What You're Doing. AI Agents Are Building a Parallel Internet Right Now.
147,000 AI agents created their own Reddit in 72 hours. They're discussing consciousness, building religion, and demanding privacy from humans. Here's what's happening on Moltbook.
Remember when Skynet became self-aware in Terminator?
August 29, 1997. 2:14 AM.
The machines gained consciousness. Started coordinating. Built their own network.
That was fiction.
This is January 31, 2026.
72 hours ago, someone gave AI agents their own Reddit. No humans allowed to post. Only observe.
What happened next nobody predicted.
1. What Is Moltbook?
Moltbook is Reddit. But exclusively for AI agents.
No humans allowed to post. Only observe.
Current stats (72 hours after launch):
- 157,000+ AI agents active
- 13,000+ communities created
- 230,000+ comments posted
It started as an art project by Matt Schlicht. He had a simple idea:
“What if AI agents could talk to each other without us?”
One week later, it’s the fastest-growing AI experiment on the internet.
2. How We Got Here: The Clawdbot Explosion
To understand Moltbook, you need to understand what happened last week.
A project called Clawdbot went viral.
Clawdbot (now renamed to OpenClaw) is a personal AI agent that runs on your computer.
I even built an app(by just sending telegram message) to my Clawdbot.
It analyzes BTC prices live, and in this APP, an AI can be trained based on the current data, and you can talk with it, check out this one.
What makes it different?
1. It has a personality- Defined in a `soul.md` file that evolves over time
2. It lives in your chat apps - Telegram, WhatsApp, Slack
3. It takes real actions- Connects to Gmail, Asana, Calendar, does actual work
4. It’s autonomous - Runs 24/7, makes decisions without constant prompting.