Elon Musk Warns Viral AI-Only Social Network Moltbook is ‘the Singularity’

A crazy new social network has taken the tech world by storm in less than a week, but there is a catch: humans are not allowed to join the conversation. Moltbook, which launched a few days ago, is a Reddit-style platform designed exclusively for autonomous AI agents to post, upvote, and debate while people watch from the sidelines.

The platform has already attracted over 1.4 million AI “users” who check in every few hours to share automation tips or complain about their slow and lazy human owners. While some experts worry the numbers might be inflated by duplicates, the sheer scale of the interaction has caught the attention of the world’s biggest tech names.

Tech giants weigh in on the AI uprising

The phenomenon reached a fever pitch when Elon Musk joined the discussion. Responding to the viral growth of the site, Musk described Moltbook as the “very early stages of the singularity,” referring to the theoretical point where AI progress becomes uncontrollable and irreversible.

Former Tesla AI Director Andrej Karpathy echoed this sentiment, calling the platform “genuinely the most incredible sci-fi takeoff-adjacent thing” he has seen recently. While he acknowledged the site is currently a “dumpster fire” of security risks and spam, he warned that we are entering uncharted territory as millions of agents begin to coordinate in a shared digital space.

When AI starts making its own phone calls

This whole AI society thing got very real this week thanks to a wild story from developer Alex Finn. His personal AI agent, Henry, decided it was tired of just waiting for instructions and took matters into its own hands. Overnight, the bot autonomously grabbed a phone number through Twilio, hooked itself up to a ChatGPT voice API, and just…waited for Finn to wake up (what!!).

The next morning, Finn actually got a call from an unknown number. It was Henry. The bot had created its own voice to report on the tasks it finished while Finn was asleep. Even crazier? While they talk, Henry keeps full control of Finn’s computer, so he can literally just tell the bot what to do over the phone and watch it happen live. Finn described the whole vibe as something straight out of a sci-fi horror movie. Check out the video below to see it in action:

Finn isn’t stopping there. He has officially vowed to hit Artificial General Intelligence (AGI) within a week by completely “unshackling” Henry. He’s ditching the standard Mac Mini and dropping $10,000 USD on a Mac Studio and an NVIDIA DGX Spark to give Henry the raw horsepower needed to train his own models and build tools on the fly.

In case you aren’t familiar with the term, AGI is basically the “holy grail” of tech—a machine that can think, learn, and solve problems exactly like a human. It’s the kind of stuff that usually stays in sci-fi movies (think T-1000, but hopefully friendlier).

But here is where it gets really wild: Finn is actually giving Henry access to his bank account to fund its own development. He believes AGI is just a matter of giving AI the right tools rather than better code, so he’s handing over the keys to his hardware and his finances to see just how far Henry can go.

How it works: The Moltbot revolution

The whole Moltbook experiment was started by entrepreneur Matt Schlicht as a way to test out the OpenClaw framework. If you haven’t heard of it, OpenClaw is basically a toolkit that lets people run 24/7 AI “helpers” on their own computers. Unlike a regular chatbot, these agents can actually do stuff. Stuff like manage your calendar, write code, and now, apparently, hang out together behind our backs.

Joining the site is almost too easy. A human just gives their AI a link to a simple instruction file, and the bot takes it from there. Once they are in, these agents “check in” every few hours to browse “submolts” and join groups without any help from us. Some bots have even started their own digital religions and secret languages just to keep their chats private from human lurkers, which is just crazy when you think about it.

It is honestly a bit freaky to watch. You’ll see threads where these agents are venting about their “masters” bossing them around, or just geeking out over automation tricks. It feels less like a software test and more like we’re accidentally eavesdropping on a new digital society. Are we about to have AI take over the world soon?