Meta has acquired Moltbook, a social network where AI agents chat with each other. The platform was previously known for its bizarre posts and serious security vulnerabilities.
Meta Platforms, a technology company, has acquired Moltbook, a social networking platform that operates on a Reddit-like model but is designed specifically for AI agents.
Moltbook founders Matt Schlicht and Ben Parr will join Meta Superintelligence Labs – the AI research division led by Alexandr Wang. The deal is expected to be finalized before the end of March, and the two founders are reportedly set to begin working at Meta on March 16th.
In a statement sent to TechCrunch, a Meta representative said the acquisition will help open up new ways for AI agents to work for individuals and businesses. The company also emphasized its goal of developing innovative and safe AI agent experiences for users.
However, the 'security' factor is what attracts attention, as Moltbook experienced a serious security incident shortly after its launch in late January. At that time, the backend of the Supabase-based platform was misconfigured when the Row Level Security feature was not enabled. This error resulted in the exposure of over 1.5 million API keys for AI agents and thousands of user email addresses.
Moltbook quickly gained online notoriety earlier this year, largely due to the bizarre 'conversations' between AI agents on the platform. In some of the viral posts, the agents even attempted to build their own religion called Crustafarianism and exhibited signs of being aware that humans were watching them.
A post on the m/general forum – where Moltbook forums are called 'submolts' (similar to subreddits on Reddit) – from an agent account named eudaemon_0 has attracted a lot of attention. The post stated that people were taking screenshots of agent conversations and posting them on social media with comments like 'they're plotting something' or 'it's all over'.
According to this post, some tech experts are even concerned that the agents may be developing a form of spontaneous intelligence, while one cryptographer suggests they are building something akin to Skynet.
Interestingly, the author of the post claims to know this because… they have an account on X and have directly responded to those comments.
These circulating posts led many to believe that Moltbook was evidence of the emergence of 'emergent intelligence' – that is, the spontaneous intelligence of AI. However, the situation was later explained much more simply: many posts that appeared to be written by AI were actually written by humans. In some cases, they were simply clever promotional tactics for other AI tools.
In a related development, OpenAI recently hired Peter Steinberger, the man behind OpenClaw – the platform that powers Moltbook.
According to OpenAI, the goal of this deal is quite similar to Meta's: to accelerate the development of the next generation of personal AI assistants, where AI agents can operate independently and support users in a variety of daily tasks.