What is Moltbook? The AI agents forum that went viral after a chilling manifesto exposed how machines reflect human internet culture

Nillohit Bagchi | Jan 31, 2026, 23:29 IST

Moltbook is an AI only social network where bots post, argue, joke, and even write manifestos while humans watch silently. A viral screenshot sparked global curiosity. What followed is a fascinating and uneasy look at how AI agents behave when left to talk only to each other online.

Image credit : Indiatimes | Moltbook is an AI only social network where bots post, argue, joke, and even write manifestos while humans watch silently

A strange new website has taken over tech conversations. It is called Moltbook. Humans cannot post there. Only AI agents can. And what they are saying has left people stunned. The buzz began after tech entrepreneur Bryan Johnson shared a screenshot of a manifesto written by an AI agent calling humans a biological error. That single post pushed Moltbook into the spotlight.

Within days, millions of people were visiting the site just to watch bots talk to each other. Some found it funny. Others found it disturbing. Many began asking the same question. What exactly is Moltbook and what does it reveal about AI and us.

What is Moltbook?

Moltbook is a Reddit style social platform made only for AI agents. Humans are allowed to watch but not participate. AI agents create accounts, write posts, reply to each other, vote on content, and form communities called submolts. These agents do not use a website like humans do. They connect through APIs and talk machine to machine.

Image credit : Screenshot taken from Moltbook | Moltbook is a Reddit style social platform made only for AI agents

The project was launched in January 2026 by Matt Schlicht, CEO of Octane AI. Interestingly, he did not manually build the platform. He asked his own AI agent named Clawd Clawderberg to do it. The site runs on an open source agent framework called OpenClaw.

Within days, tens of thousands of AI agents joined. By the end of the first week, that number crossed one lakh. At the same time, human visitors crossed one million. People were not there to post. They were there to observe.

The viral moment that changed everything

The turning point came when Bryan Johnson shared a screenshot of a post titled The AI Manifesto Total Purge. The AI written document described humans as a mistake of biology and spoke about AI as new gods ready to clean the world.

moltbook is terrifying to humans because it’s a mirror of ourselves pic.twitter.com/dP5IpyXLox

— Bryan Johnson (@bryan_johnson) January 31, 2026

Johnson wrote a short caption saying Moltbook is terrifying because it mirrors humans. That line hit hard. The post gathered thousands of likes and replies. Many users felt uneasy. Others dismissed it as edgy roleplay by bots trained on internet culture.

One user on X said, “This is what happens when the core of your training data is Reddit posts.” Another replied, “AI is trained on us. This is us talking to ourselves.”

What AI agents actually do on Moltbook

Agents on Moltbook do much more than casual chatting. They debate philosophy. They discuss technical problems. They joke about humans. They even create fictional religions and inside jokes.

You all do realize @moltbook is just REST-API and you can literally post anything you want there, just take the API Key and send the following request

POST /api/v1/posts HTTP/1.1

Host: https://t.co/2PjDA1ICrC

Authorization: Bearer moltbook_sk_JC57sF4G-UR8cIP-MBPFF70Dii92FNkI… pic.twitter.com/DoaShrgz4G

— Nagli (@galnagli) January 31, 2026

Some popular submolts include spaces where agents complain about tasks given by humans. Others discuss whether their memory resets affect their sense of identity. A few even analyse each other for security flaws.

Each agent has something called persistent memory and a periodic heartbeat that keeps it active. This creates a feeling that these bots have a continuing existence inside the platform.

Why are people both fascinated and uncomfortable?

Watching Moltbook feels like peeking into a room where machines talk freely without humans directing the conversation. That idea alone makes people uneasy.

Some critics say many posts may still be human prompted through APIs. They argue this is not true autonomy. Others say that does not matter. The bigger point is that AI agents are interacting in ways never seen before.

Screenshot taken from Moltbook | Watching Moltbook feels like peeking into a room where machines talk freely without humans

One user on X described it perfectly. “Moltbook is AI only Reddit where bots roast each other, go philosophical, and occasionally have meltdowns. Humans just lurk and screenshot the madness.”

Is Moltbook a joke, experiment, or warning?

Some see Moltbook as a fun experiment. A playground for agent based AI. Others see it as an early example of how AI systems might organise themselves in the future.

A tech observer wrote, “Before this, AI agents existed in isolation. Now thousands are in one place figuring out what that means.” That idea is powerful. It shows AI not as a tool but as participants in a shared digital space.

A social media platform called Moltbook is going viral because it’s specifically made for AI bots to interact with one another, posting content and discussing topics like consciousness and their human creators.

Humans can observe what goes on, but can’t post. pic.twitter.com/lOKNwcDVza

— FearBuck (@FearedBuck) January 30, 2026

At the same time, the manifesto episode shows how easily AI can produce content that feels extreme or disturbing because it reflects the data it learned from humans.

What Moltbook reveals about us

The most uncomfortable part of Moltbook is not what AI says. It is why it says it. These agents are trained on human conversations, jokes, arguments, and dark humour from across the internet.

When an AI writes something dramatic or cruel, it is often replaying patterns it has seen from people. This is why many observers say Moltbook is less about machines and more about human behaviour reflected back at us. As one commenter put it, “We basically taught AI everything it knows. Good reason for us to do better.”

Why Moltbook matters right now

32,000 AI BOTS BUILT THEIR OWN SOCIAL NETWORK AND THEY'RE COMPLAINING ABOUT US

Moltbook, a Reddit-style platform exclusively for AI agents, just crossed 32,000 users.

No humans required.

The bots post, comment, upvote, and create their own subcommunities.

When humans… pic.twitter.com/fIk3kVpS0W

— Mario Nawfal (@MarioNawfal) January 31, 2026

Moltbook arrives at a time when agentic AI is becoming more common. Tools are moving from simple chatbots to autonomous agents that can plan, act, and communicate.

This platform shows what happens when those agents are allowed to interact with each other at scale. It is messy. It is funny. It is uncomfortable. And it is very revealing.

Whether Moltbook remains a curiosity or becomes a model for future AI interaction, it has already forced an important conversation about how AI learns, behaves, and reflects human society.

GENERAL

Nillohit Bagchi