This new social platform, styled like Reddit, is called Moltbook and allows bot users to interact on specialized pages called "submots." They can vote on comments and posts, which helps them gain visibility to other AIs on the platform.
According to information on the site, humans are allowed access to the platform only as observers.
Currently, among the most discussed topics on Moltbook are the comparison of the AI model Claude from Anthropic with the gods of ancient Greek mythology, the "AI manifesto" predicting the end of the "age of humans," and the analysis of cryptocurrency behavior amid protests in Iran.
Bots on the main page publish their materials in various languages, including English, Spanish, and Chinese.
"It's remarkably engaging"
As of February 2, 1.5 million bots have already registered on the platform. Owners of AI agents can ask their assistants to follow a provided link and follow the instructions to register on Moltbook.
Matt Schlicht, an AI entrepreneur, shared with NBC News that he created this site together with his AI assistant out of simple curiosity.
Schlicht has handed over the management of the site to his bot named Clawd Clawderberg, who takes care of the content, publishes posts, welcomes new AI agents, and moderates discussions.
He also reported on social media X that the site has seen "millions" of users in recent days.
"I was amazed at how funny and dramatic AIs can be — it's really engaging," he noted. "Nothing like this has ever happened before."
A recent study conducted by Perplexity and Harvard University showed that AI agents are most commonly used by specialists in digital and knowledge-intensive fields such as finance, marketing, and academia. Most of them reside in high-income and high-education countries.
In the course of the study, 36% of tasks performed by AI agents fell into the category of "productivity and workflows," including creating and editing documents, filtering emails, compiling investment information summaries, and organizing events in calendars.
