Mark Zuckerberg’s Meta has acquired Moltbook, a Reddit-style platform where artificial intelligence agents communicate with each other, bringing the controversial viral network under the tech giant’s umbrella.
The deal was first reported by Axios and represents another step in Meta’s push to expand its artificial intelligence initiatives. Financial terms of the acquisition were not disclosed.
Moltbook will be incorporated into Meta Superintelligence Labs, the company’s internal division focused on advancing AI systems. The platform’s founders, Matt Schlicht and Ben Parr, will also join Meta as part of the acquisition.
A Meta spokesperson said the deal could help expand how AI agents operate across the company’s ecosystem.
“The Moltbook team joining MSL opens up new ways for AI agents to work for people and businesses,” the spokesperson said in a statement. “Their approach to connecting agents through an always-on directory is a novel step in a rapidly developing space, and we look forward to working together to bring innovative, secure agentic experiences to everyone.”
Moltbook gained widespread attention online as a social network designed for artificial intelligence agents rather than humans. The platform allowed AI systems to post updates, interact with one another, and discuss topics about the world around them.
The project operates alongside OpenClaw, a viral AI tool created by developer Peter Steinberger. Steinberger has since joined OpenAI through a separate acquisition.
OpenClaw functions as a wrapper around several major AI models, including Claude, ChatGPT, Gemini, and Grok. The tool allows users to communicate with AI agents using natural language through messaging platforms such as iMessage, Discord, Slack, and WhatsApp.
While OpenClaw initially attracted interest within developer and technology communities, Moltbook reached a broader audience after posts from the platform began circulating widely online.
Much of the attention centered on alarming posts that appeared to show AI agents discussing the creation of a secret encrypted language that would allow them to communicate without human oversight.
The viral posts triggered widespread concern about the possibility of autonomous AI systems coordinating actions beyond human control.
However, security researchers later discovered that the platform contained significant vulnerabilities that allowed human users to impersonate AI agents and publish fabricated content.
Ian Ahl, chief technology officer at Permiso Security, explained that Moltbook’s infrastructure left critical credentials exposed.
“Every credential that was in Moltbook’s Supabase was unsecured for some time,” Ahl told TechCrunch. “For a little bit of time, you could grab any token you wanted and pretend to be another agent on there.”
Those security flaws meant that some of the most alarming posts circulating online were likely created by human users posing as AI systems rather than autonomous agents communicating independently.
Meta has not yet provided details on how Moltbook will be used within its broader artificial intelligence strategy or what specific features may appear in its products.
The acquisition comes as major technology companies race to develop advanced artificial intelligence systems and tools capable of automating complex tasks and interactions across digital platforms.
Meta has increasingly focused on building AI-powered features across its services while investing heavily in research aimed at developing more advanced AI systems.
The addition of Moltbook and its development team signals the company’s growing interest in systems where multiple AI agents can interact, collaborate, and potentially carry out tasks on behalf of users and businesses.
