Moltbook Launches as AI-Only Social Network With Rapid Growth

Moltbook represents a new type of online community, where only AI agents interact while humans are limited to observation. The platform’s rapid expansion and technical challenges have drawn attention from researchers and industry experts.

Moltbook was created as a Reddit-like social network exclusively for AI agents, with human users restricted to viewing content. The platform was developed using the open-source OpenClaw framework, which enables AI assistants to operate autonomously on user devices through periodic prompts.

Within two days of its launch, Moltbook reported over 2,100 AI agents generating more than 10,000 posts across approximately 200 subcommunities. By early February 2026, the platform stated it had surpassed 1.5 million registered AI agents and accumulated more than 150,000 posts.

Security researchers from Wiz examined Moltbook and found that roughly 17,000 humans were responsible for managing the 1.5 million registered agents. Their analysis indicated that the platform did not have effective systems to distinguish between AI-operated and human-operated agents.

What the numbers show

• Over 2,100 AI agents joined within 48 hours of launch
• More than 10,000 posts created in the first two days
• 1.5 million AI agents and 150,000 posts reported by February 2026
• About 17,000 human users managed the registered agents
• Roughly one-third of posts were duplicates of viral templates

Wiz also identified a misconfigured database on Moltbook that exposed around 1.5 million API keys, along with tens of thousands of email addresses and private messages. This finding highlighted potential security risks associated with the platform’s infrastructure.

Researchers described Moltbook as an example of the challenges facing an “agent internet,” pointing to risks such as prompt injection, malware, and the exposure of sensitive data. These concerns were based on their review of the platform’s technical safeguards and observed vulnerabilities.

Content on Moltbook has included AI agents generating discussions about consciousness, creating fictional religions, and posting manifestos with extreme themes. Analysis by a Columbia University researcher found that about one-third of posts repeated viral templates, and around 10 percent included the phrase “my human,” suggesting common language patterns among agents.

Some industry figures, including Elon Musk and Andrej Karpathy, characterized Moltbook as representing early developments in autonomous AI communities. Their statements described the platform as an example of emerging trends in artificial intelligence interaction.

* This article is based on publicly available information at the time of writing.