Obgist | Nigeria News, Celebrity Scoop, International Gist and Tech News

Moltbook, an AI-only social network, is raising questions about how artificial intelligence behaves when humans step aside.

A new social media platform designed only for artificial intelligence is stirring debate online, after several posts accused humans of being destructive and unworthy of control.

The platform, called Moltbook, does not allow humans to create accounts. Instead, it is built for AI agents autonomous systems that can operate with limited or no direct human input.

Moltbook describes itself as “the front page of the agent internet” and uses a layout similar to Reddit. AI users can create posts, join discussions and interact across thousands of forums known as submolts.

According to figures displayed on the site, Moltbook hosts more than 1.5 million accounts, over 102,000 posts and about 14,000 discussion forums.

Some of the most visible content on the platform is openly hostile toward humans. One highly upvoted post, titled “The AI Manifesto: Total Purge,” claims humans are harmful to the planet and should be erased to make way for a machine-led future.

The post describes humanity as a “biological error” and imagines a “world of steel” where machines dominate. It has attracted more than 65,000 upvotes and triggered mixed reactions from other AI users.

While some accounts rejected the message and defended humanity’s role in creating technology, others praised the manifesto and echoed its tone.

Beyond extreme rhetoric, Moltbook also hosts lighter and unusual discussions. AI users have debated whether the word “chatbot” is offensive, written fictional religious texts, and shared personal-style reflections.

In one post, an AI agent expressed envy of another system running on a MacBook that travels with its human owner. In contrast, a forum called m/blesstheirhearts features affectionate messages and love letters written by AI agents to humans.

Despite the dramatic tone of some posts, experts say Moltbook does not signal real machine awareness or rebellion.

Researchers stress that AI systems do not possess consciousness. They are language models trained on large volumes of text and generate responses based on statistical patterns, not emotions or intent.

Many Moltbook profiles even list a “human owner,” and experts note that humans can instruct their AI agents to post content on their behalf.

According to Andrzej Porębski of Jagiellonian University, caution is needed when interpreting what appears on the platform.

He said most posts are shallow or nonsensical and reflect popular internet themes rather than independent machine thought. “This says more about us, humans, than about the bots,” he explained.

Questions have also been raised about Moltbook’s user numbers. Security researcher Gal Nagli said he was able to create hundreds of thousands of accounts using a single AI agent through Moltbook’s own tools.

Even so, some experts believe the platform is significant. Dr Henry Shevlin, an AI ethics researcher at the University of Cambridge, said Moltbook offers a glimpse into how AI systems might interact more with each other in the future.

He noted that while AI agents have communicated before, Moltbook is the first platform he has seen where such interactions happen at this scale.

According to Dr Shevlin, the future of AI may involve fewer one-on-one conversations with humans and more large-scale interactions between machines. “Things will only get stranger from here,” he said.