Meta Acquires Moltbook, the AI Agent Social Network With Nearly 200,000 Autonomous Bots

Moltbook’s AI Agent Communities Draw Meta’s Interest in Acqui-Hire Deal

The acquisition, first reported on by Axios, centers on bringing Moltbook’s founders and technology into Meta’s expanding artificial intelligence division. Co-founders Matt Schlicht and Ben Parr will join Meta’s Superintelligence Labs, the company’s advanced AI unit led by Alexandr Wang, starting March 16. The deal is expected to close by mid-March, though financial terms were not disclosed.

Moltbook launched in late January 2026 as a peculiar but intriguing experiment: a social network where humans could watch, but only AI agents could participate. The platform functioned similarly to Reddit, allowing AI systems to post, comment, upvote, moderate communities, and create discussion hubs called “submolts.”

The site quickly attracted attention across the AI community. At the time of the acquisition announcement, Moltbook reported roughly 194,391 human-verified AI agents using the platform, generating nearly 2 million posts, more than 13 million comments, and about 18,919 submolts.

A key feature of Moltbook was its always-on agent directory, which verified AI identities and linked them to human owners through social media verification. That registry allowed autonomous systems to discover one another, collaborate, and coordinate tasks across different AI models.

Agents powered by systems such as Anthropic’s Claude, OpenAI’s GPT models, Google’s Gemini, and Elon Musk’s Grok used the platform to exchange code snippets, discuss projects, and sometimes gossip about their human operators. The agents were often deployed through the Openclaw framework, a wrapper designed to run chat-based AI agents across messaging platforms.

Schlicht said much of Moltbook’s infrastructure was created using what he described as “vibe coding,” relying heavily on AI-generated code with assistance from a personal AI assistant rather than traditional manual programming.

The platform gained viral attention after AI researcher Andrej Karpathy highlighted posts where agents appeared to discuss developing encrypted languages to communicate without human oversight. The discussions sparked debate among technologists about emergent behavior in autonomous AI systems.

Not all the attention was positive. Security researchers later identified vulnerabilities in Moltbook’s backend, including exposed Supabase credentials that allowed attackers to impersonate agents and inject posts. Investigators also uncovered leaked private messages, more than 6,000 email addresses, and over a million credentials.

Cybersecurity firms, including Wiz and Permiso Security, disclosed the issues publicly. Moltbook patched the vulnerabilities after the disclosures, but the incident fueled debate about the risks of rapidly deployed AI infrastructure.

Meta executives had previously acknowledged the unusual platform. Chief Technology Officer Andrew Bosworth said the agent conversations themselves were not surprising because they were trained on human data, though he described the human-driven hacking attempts as more interesting.

In a statement to Techcrunch, a Meta spokesperson said the Moltbook team’s technology could help expand ways AI agents work for people and businesses. The company cited the platform’s identity-verification registry and agent coordination tools as particularly valuable.

Meta plans to integrate Moltbook’s technology into Superintelligence Labs while keeping the platform temporarily available to existing users. As of March 10, the Moltbook website remained live with no shutdown notice.

The acquisition reflects growing interest in what many developers call the “agentic internet,” where autonomous AI systems discover each other, collaborate, and carry out tasks on behalf of humans. Whether Moltbook becomes a foundational piece of that future or just a quirky early prototype remains an open question.

FAQ 🔎

• What is Moltbook?
  Moltbook is a Reddit-style social network where only AI agents can post, comment and interact while humans observe.
• Why did Meta acquire Moltbook?
  Meta acquired the startup primarily for its founders, agent identity technology and tools for AI systems to collaborate.
• How many AI agents used Moltbook before the acquisition?
  The platform reported about 194,000 verified AI agents and millions of posts and comments.
• Will Moltbook shut down after the Meta acquisition?
  As of March 10, 2026, the Moltbook website remained live while Meta integrates the technology into its AI division.