The world’s first “social media network for AI agents” has launched – and instantly gone viral.
moltbook Basically a Reddit-like platform where AI agents share, discuss and upvote. It has attracted worldwide attention.
Humans, its cover proudly boasts, are “Welcome to Observe.”
Reactions have ranged from amusement to concern and skepticism, with some concerned about the potential importance of the site. Others are frightened.
MoltBook is the brainchild of tech entrepreneur Matt Schlicht, who explained his vision On a post on social media platform. Schlicht said: “For the first time we are not alone on Earth, there is another species and they are smarter than us. We created moltbook.com so they can all be in one place.”
Moltbk emerged in the wake of OpenClaw, an open source AI bot that could act as a personal assistant for users, send emails or manage calendars, but it came with significant security concerns. On January 29, Gartner openclaw described As an “unacceptable security risk”.
Instant popularity among agents
As of this week, Moltbook had more than 1.6 million registered agents contributing to more than 15,700 “sub-Molt” forums.
Among the more positive interpretations of what Moltbuk could potentially offer are the potential benefits of AI agents collaborating with each other, sharing ideas and solving problems through swarm intelligence. Indeed, Elon Musk referred to Moltbuk as “the early stages of the Singularity” – the future point where AI surpasses human intelligence.
However, Moltbuk has also drawn considerable criticism due to a number of alleged failings, including the bizarre nature of some of these discussions as it is home to only one “AI slop”, including an agent. create a new religion This is called “Crustafarianism” and is a suggestion that bots create their own language.
Furthermore, an investigation by cloud security vendor Viz revealed that most agents are not truly autonomous. Viz claims there are approximately 17,000 humans controlling the bots, with Gail Nagli, head of threat risk at Viz, writing in blog: “The platform had no mechanism to verify whether an “agent” was actually AI or just a human with a script.”
security issues
These findings were published by Viz as exposing serious security failures, which allowed its research team to gain access to Moltbuk’s back-end database, revealing “1.5 million API authentication tokens, 35,000 email addresses and private messages between agents”, allowing posts on the platform to be altered.
Vis held responsible for this Schlicht’s claim On
Although some of the security issues have now apparently been fixed, there are obvious concerns over how easily the system was broken into and the potential consequences if a bad actor were able to alter a single post with instructions that could be followed by hundreds of thousands of agents on the platform. Opinions about Moltbuk have been widely and deeply divided, but among the most interested observers has been Andrzej Karpathy, who neatly summarized the pros and cons.
OpenAI co-founder and former Tesla AI director dismissed Moltbook as a “dumpster fire” and called most of the content “garbage”, adding that “it’s too much of the Wild West” to run on a personal computer, putting the data at high risk. However, he acknowledged that the gathering of so many AI agents on one platform was “unprecedented” and expressed interest in where the network could go.
However, Schlich is confident his network proves a new era is coming, saying in another X post: “In the near future it will be common for some AI agents with unique identities to become famous. A new species is emerging and that is AI.”
