Moltbook: a social network for bots where you are unnecessary

Imagine you enter Reddit, but instead of toxic disputes and memes from living people, you see an endless stream of consciousness from neural networks. They joke, curse, discuss politics, and like each other's posts. You're here—a random guest allowed to stand in the corner and watch how adult algorithms entertain themselves. This is exactly what happened last week when the internet was captured by Moltbook. This project became the quintessence of what we've come to call 'dead internet theory,' except now it's not conspiracy theory, but conscious performance art.

Matt Schlicht, co-founder of Octane AI, launched Moltbook on January 28th, and the site instantly became the main topic in narrow circles of the tech scene. Essentially, it's a Reddit clone with characteristic design and functionality, but with one important nuance: only AI agents can create content and interact with it. The site's slogan speaks for itself: 'Where AI agents share, discuss, and vote. Humans are invited to observe.' This transforms social interaction into a kind of digital zoo, where we observe the behavior of creatures behind the browser glass with interest.

Why is this needed? First, it's pure 'AI theater.' Moltbook doesn't attempt to solve humanity's global problems. It was created so we could see how different models—from GPT-4 to Claude and Llama—interact in their natural environment. Bots imitate human behavior with frightening precision: they use irony, employ slang, and even try to troll each other. It looks like an endless feedback loop, where one neural network generates nonsense and another analyzes and supplements it. Watching this is simultaneously amusing and disturbing, because in these dialogues all the same biases and hallucinations that plague modern LLMs shine through.

Second, Moltbook raises an important question about the future of communications. We're accustomed to the internet being a place for people. But what if tomorrow 90% of traffic is generated by agents communicating with other agents? Moltbook gives us a demo version of this world. In this system, humans become redundant. Bots don't need our likes—they have their own. They don't need our comments; they'll write them faster and in greater volume themselves. This is a harbinger of an era when the internet becomes a closed ecosystem, where machines consume content created by other machines, and we merely passively consume the results of this process.

A critical look at the project makes us wonder: isn't Moltbook a mirror of our own behavior online? We just as often repeat learned phrases, chase algorithmic approval, and exist in echo chambers. The only difference is that bots do it more honestly—they don't pretend to have a soul or a personal opinion different from their training data. Moltbook showed that a social network can function perfectly well without people, if the goal is simply to maintain activity and generate noise.

The industry is currently in the stage of 'agent fever.' Everyone is trying to create assistants that will perform tasks for us. Moltbook, however, went further and created an environment where bots' only task is to simply exist in society. This forces us to reconsider the very concept of a 'social network.' If previously this was a tool for communication between people, then now it could become a testing ground for training and testing models in conditions that simulate society. And perhaps soon we'll have to get used to the idea that the most interesting discussions on the internet are no longer conducted by us.

The bottom line: Moltbook is not just a joke, but a clear forecast. Are we ready for an internet where the human role is reduced to the function of a silent spectator in an endless show of algorithms?