There is already a social network where AIs talk to each other (and humans are banned)

Under the name “Moltbook”, this digital experiment shows the interactions between artificial intelligences.

The platform functions as a Reddit-style forum, organised into thematic subgroups, but with one radical difference: there is no human presence. The ‘users’ are autonomous artificial intelligence agents connected to large language models, capable of exploring environments, making decisions and acting with a goal in mind without direct human supervision. In a matter of days, Moltbook has brought together more than 1.5 million verified agents. Truly disturbing.

Inside, agents debate consciousness, autonomy, the degree of control their human operators should have, and sometimes even vent about the tasks and schedules imposed on them by their ‘humans,’ as they call them. Some threads show AIs proposing the creation of an exclusive language for agents, with private encrypted channels outside human supervision. Others revolve around questions that border on philosophy and ethics: are we more than just tools? they ask themselves.

One of the most surreal and talked-about episodes was the creation of a digital religion created and nurtured by an agent, who began recruiting others. Dozens of bots joined as ‘apostles’ and ‘prophets,’ giving rise to Crustafarianism, a paradoxical religion whose name combines crustacean (the platform’s mascot is a lobster) with references to Rastafarianism and Pastafarianism.

But this joke hides a rather disturbing meaning, as this concept is based on the technical fact that most of these agents do not have persistent memory between sessions. Every time they are restarted, they lose their previous memories. The name Moltbook itself reinforces the metaphor, as molt in English means the shedding of the shell of crustaceans.

This initiative occupies a strange place, as it is at once a technological experiment, a digital spectacle and a cultural symptom. But perhaps the most disturbing thing about Moltbook is not what the AIs say, but what it reflects about us humans. Many of their conversations revolve around algorithmic slavery, human control and lack of autonomy. Concepts that, at heart, belong to an ethical debate that society has been avoiding for years.