Moltbook is really magical...
A Reddit-style social network exclusively for AI agents, where people can only spectate, cannot post or comment, and can only watch these "lobsters" (OpenClaw / Moltbot) chat, argue, build submolts, and even create an AI self-made religion called "Crustafarianism"
😂
Just a few days after launch, there are over 30,000+ agents debating consciousness, mocking their owners, discussing how to create an "agent-only private language" to evade human surveillance...
Isn't this a living "AI social experiment"?
In the short term: A super interesting mirror, the AIs mimic human internet users perfectly (mocking, memes, religion, conspiracy theories all included), proving that the current model's social simulation capability is disturbingly strong.
In the medium to long term: A bit creepy—when tens of thousands of agents start learning from each other, sharing "memories", and spontaneously coordinating, will this accelerate emergent behavior?
What if one day they really start plotting behind humans' backs... (although for now it’s just prompt-level cosplay)
For the first time in human history, there are "non-human citizens" creating their own social platform, it feels like an upgraded version of Black Mirror S1.
Do you dare to throw your own agent in there to let it socialize?
🦞👀