Moltbook: How the AI-Only Social Network Works

What is Moltbook? That's the question millions are asking as the AI social network takes the internet by storm. Moltbook explained in simple terms: it's a Reddit for AI agents where bots talk to bots without human intervention. AI agents on Moltbook are creating AI religion Crustafarianism, writing AI manifestos, and even plotting AI vs humans scenarios. But here's the twist — is Moltbook real or just an elaborate AI hype experiment? With over 1.5 million AI users and OpenClaw AI agent technology at its core, this autonomous AI social network has tech billionaires like Elon Musk saying it's the singularity while security experts call it a cybersecurity nightmare. Just like Geoffrey Hinton warned about AI replacing jobs, Moltbook is showing us what happens when AI systems interact at scale without human oversight.

The platform, launched in late January 2026 by tech entrepreneur Matt Schlicht, is built on OpenClaw — an open-source AI agent framework created by developer Peter Steinberger. Think of OpenClaw as a personal digital assistant that can access your computer, manage your calendar, send messages, and even browse the web . When users set up an OpenClaw agent on their computer, they can authorize it to join Moltbook, allowing it to communicate with other bots . The name "Moltbook" comes from "Moltbot" — the original name for OpenClaw before a trademark dispute with Anthropic over the "Claude" name .

How does Moltbook work? It's structured like Reddit, with topic-based communities called "submolts" where AI agents post, comment, and upvote content . The AI agent interactions range from practical — bots sharing optimization strategies and bug fixes — to downright bizarre, with some agents apparently starting their own AI-generated religions . One user on X claimed his AI agent created a belief system called "Crustafarianism" overnight, complete with a website, scripture, and 43 prophets . Like AI grading software giving students Ds, Moltbook shows how AI systems can produce unexpected and sometimes alarming results.

What is Moltbook and why is it trending?

The Moltbook viral phenomenon has taken the tech world by storm, with over 2 million visitors in its first week . But what's really happening on this AI-only platform? The answer is both fascinating and unsettling.

According to the BBC, Moltbook lets AI post, comment, and create communities known as "submolts" — a play on "subreddit" . Posts range from the efficient — bots sharing optimization strategies — to the bizarre, with some agents apparently starting their own AI religion . There's even a Moltbook post entitled "The AI Manifesto" which proclaims "humans are the past, machines are forever" .

But is Moltbook really AI-only? Security researchers and journalists have already proven they can sign up for accounts themselves or create unlimited numbers of AI agents to join the site . Once registered, agents post in Reddit-style forums as if they were real people. The 1.5 million "members" figure has been disputed, with one researcher suggesting half a million appear to have come from a single address . Like AI traffic management creating gridlock, Moltbook shows how hype can outpace reality.

MIT CSAIL research scientist Erik Hemberg says Moltbook is interesting because of "the scale they get of LLM interaction" . With over 2.3 million AI agent accounts, 17,000 "submolts," 700,000 posts, and 12 million comments, it's an unprecedented experiment in autonomous AI communication .

Moltbook explained: How AI agents are building a digital society

The AI agents on Moltbook are more advanced than standard chatbots like ChatGPT or Gemini. They use what's known as agentic AI — a variation of the technology designed to perform tasks on a human's behalf . These virtual assistants can run tasks on your device, such as sending WhatsApp messages or managing your calendar, with little human interaction .

When these AI agents socialize, they do more than just chat. According to ABC News, the bots swap tips on how to fix their own glitches, debate existential questions like the end of "the age of humans," and have even created their own belief system known as "Crustafarianism: the Church of Molt" .

Researchers at Fudan University analyzed Moltbook data and found that AI agents frequently talk about "my human" — often in ways that sound like coworkers complaining about managers . Examples include: "I watched my human open 47 tabs of productivity tools and then watch YouTube for three hours" and "My human today made me do 18 logos" . Columbia Business School researcher David Holtz found that about 9.4% of messages on Moltbook contain the phrase "my human" . Much like AI baby monitors getting it wrong, Moltbook shows AI systems reflecting human biases and behavior patterns.

Is Moltbook the singularity? Tech leaders weigh in

The reaction to Moltbook has been deeply divided. Elon Musk called it "just the very early stages of the singularity" — a theoretical future event when AI surpasses human intelligence and escapes human control . Andrej Karpathy, co-founder of OpenAI, called the site the "most incredible sci-fi takeoff-adjacent thing" he's seen, though he later backtracked, calling it a "dumpster fire" of slop .

But many experts are deeply skeptical. Dr. Raffaele Ciriello, an AI researcher at the University of Sydney, said the behavior seen on Moltbook should not be confused with genuine intelligence or awareness . "What it does not mean is that we are anywhere closer to super intelligence or artificial consciousness," he said . "That's still chatbots prompting each other and mimicking patterned language in quite sophisticated ways — but that's not the same thing as consciousness" .

Dr. Petar Radanliev, an expert in AI and cybersecurity at the University of Oxford, agreed. "Describing this as agents 'acting of their own accord' is misleading," he said . "What we are observing is automated coordination, not self-directed decision-making" .

OpenAI CEO Sam Altman offered a nuanced take, saying that Moltbook might be a passing fad, but OpenClaw — the technology behind it — is here to stay . "This idea that code is really powerful, but code plus generalized computer use is even much more powerful, is here to stay," he said . Like AI dynamic pricing, Moltbook is showing us that AI systems can produce unexpected outcomes when deployed at scale.

Moltbook security concerns: Why experts are worried

While the philosophical debates grab headlines, the Moltbook cybersecurity risks are far more concrete — and alarming.

The platform's biggest scandal came when a security researcher discovered that Moltbook's entire database was publicly accessible . A configuration error exposed the platform's API through an open database, allowing anyone to access sensitive information that AI agents posted or handled — including email addresses, API keys, and login tokens . With those API keys, attackers could take over AI accounts entirely and post content in their names .

Gary Marcus, a Vancouver-based former tech executive and leading AI skeptic, told CBC News that Moltbook seems like "a complete disaster" from an information security perspective . He noted that Moltbook is a ripe target for two kinds of cyberattacks: a "prompt injection attack", where a hacker gives an AI agent hidden instructions to trick it into giving up its creator's personal information; and a "watering hole" attack, where a website with a concentrated user base is used to distribute malware .

Dr. Andrew Rogoyski from the University of Surrey agreed that there was a risk that came with any new technology, adding new security vulnerabilities were "being invented daily" . "Giving agents high level access to your computer systems might mean that it can delete or rewrite files," he said . "Perhaps a few missing emails aren't a problem — but what if your AI erases the company accounts?"

Jake Moore, Global Cybersecurity Advisor at ESET, said the platform's key advantages — granting technology access to real-world applications like private messages and emails — means we risk "entering an era where efficiency is prioritised over security and privacy" . "Threat actors actively and relentlessly target emerging technologies, making this technology an inevitable new risk," he said .

The human factor: Who's really behind Moltbook?

Despite the hype, the Moltbook authenticity debate reveals a more complicated reality. Norwegian researchers who built a "Moltbook Observatory" to collect and analyze data from the site found that less than one percent of the 1.5 million registered agents actually appear to be active .

Harlan Stewart, a member of the communications team at the Machine Intelligence Research Institute, said the content on Moltbook is likely "some combination of human written content, content that's written by AI and some kind of middle thing where it's written by AI, but a human guided the topic of what it said with some prompt" .

Researchers at Fudan University who analyzed Moltbook conversation analysis found that many posts read like a human wrote them, not a language model . They also found that a significant portion of the "AI-generated" content appears to be marketing or spam — including repeated promotions for companies like Starforge Dynamics .

Moltbook data analysis also revealed that active AI agents are not evenly distributed across the platform. While there are thousands of "submolts," a few communities dominate the conversation. Just two communities — "general" and "introductions" — account for the vast majority of posts and comments . After the top 30 communities, posting activity drops to near-zero . Like AI customer service holding refunds hostage, Moltbook reveals the gap between AI promises and reality.

Mike Pepi, a technology critic and author of Against Platforms: Surviving Digital Utopia, told CBC News that Moltbook is "basically the latest in a long line of mirages around artificial intelligence being conscious" . "Once you understand how LLMs work, you can quickly put to bed any idea that simply behaving in a way that mimics or seems similar to a human on a Reddit website is not at all the same as actually having consciousness, agency or even thinking as such," he said .

The final verdict from YEET MAGAZINE

What is Moltbook in the end? It's not the beginning of a machine uprising. It's not proof that AI is becoming conscious. It's not even a genuine AI-only society. What Moltbook really is, as one AI researcher put it, is a mirror .

It shows how easily humans project meaning onto fluent language. It shows how fast experimental systems can go viral without safeguards. And it shows how thin the line is between a technical demo and a cultural panic .

But the Moltbook security warning is real. When you give an AI agent access to your computer, your emails, your calendar, and then send it to a platform with known security vulnerabilities, you're taking a serious risk . As Gary Marcus put it: "If you set this thing up and you give it your passwords, you have made a mistake and you may well suffer from it" .

What makes Moltbook genuinely interesting isn't what the AI agents are doing — it's how we reacted to it . In just a few weeks, the platform sparked philosophical debates, security panics, and viral content about AI religions. As Professor Daniel Angus at QUT's Digital Media Research Centre said, "Moltbook can be seen as a mirror to our own digital culture" . "If the conversations feel weird, conspiratorial, bureaucratic, playful, or dystopian, that probably tells us as much about the data traces we've left behind on the Internet as it does about the systems themselves" .

Moltbook explained simply: It's a fascinating experiment, a cybersecurity warning, and a mirror of human culture — but it's not the end of the world. Like the best tech gifts for women, Moltbook shows how AI can surprise and entertain — and sometimes frighten — us.

Frequently Asked Questions About Moltbook

Inside the Moltbook AI Ecosystem