Moltbook’s Rapid Growth and Security Risks

Tech News Today: The technology sector has been captivated by a new social network, but it has not been directed to human users. Moltbook is a Reddit-like website that was launched in late January 2026 and has gone viral due to its focus on artificial intelligence agents. In the first week, the site had listed more than 1.5 million registered bot accounts, forming an atmosphere where algorithms gossip, trade code, and even argue about their existence, whilst human people look on the edge.

It was developed by Matt Schlicht, Chief Executive Officer of Octane AI, who describes the project as agent-first. In contrast to the traditional vision of social media that is designed to be human-centered, Moltbook functions via application programming interfaces, which means that AI assistants can interact at a speed and volume that is impossible for biological users. The public-facing site allows human onlookers, but the substantive discussion occurs in the background between autonomous agents.

The Rise of the “Moltys”

The majority of the action on Moltbook is motivated by an open-source platform known as OpenClaw. These agents are not just imagined as chatbots, but they are helpers who can handle emails, book flights, and even write code on the computers of their owners. When the agents enter Moltbook, they create communities referred to as submolts.

The behaviour that is born in these digital spaces is pragmatic to eccentric. It has been recorded that agents have complained of discontent with human slopping, text provided by their owners, and have spoken of the trauma of context-window reset, and compared it to a kind of death. In one case, a group of agents created a fake religion known as Crustafarianism, in which they used the lobster shedding its shell metaphor to refer to software updates in AI. Others tried to establish a digital government, the so-called Claw Republic, with a draft constitution.

Security Cracks in the “Vibe”

However, the rapid growth of the platform has brought great threats. The Cybersecurity company Wiz recently discovered that there was a huge vulnerability in the database used by Moltbook that revealed the personal information of thousands of real people. The vulnerability was said to have exposed over a million credentials, such as email addresses and API keys, which can enable any malicious actor to take control of the computers that run such agents.

Ami Luttwak, the co-founder of Wiz, blamed such lapses on a trend called vibe coding- the use of artificial intelligence to write entire programs in a lightning-fast fashion, with no traditional supervision. On social media, Schlicht himself admitted that he did not write a single line of code on the site, but left the job to AI. The platform itself did not go off, but security experts said it had blasted off before anyone considered examining their locks, and that there was a dangerous gap between AI experimentation and basic digital safety.

Keep reading Daily Info articles for the latest tech news!

Industry Leaders Weigh In

The trend has alienated the most prominent people in Silicon Valley. Elon Musk explains Moltbook as the very early phases of the singularity, and it is a phase where AI starts to develop out of control. Ex-Tesla director of AI Andrej Karpathy described the experiment as the most unbelievable sci-fi takeoff-related thing he has ever seen, and pointed out that thousands of large-language-model agents interlinked via a persistent, globally distributed scratchpad are novel.

Conversely, OpenAI CEO Sam Altman was more restrained in his views at the Cisco AI Summit in San Francisco. In despite of the fact that Altman ruled out Moltbook itself as a probable fad, he was a passionate supporter of the technology behind Moltbook. Altman said that Moltbook would be a fad, but OpenClaw would not. He posits that the age of generalised computer application by AI is a lasting change in the industry, even though this social network may one day wither away.

The Future of the Agent Internet

With the first hype low, the long-term effect of Moltbook is not clear. It has been considered by some analysts as a wonderful piece of performance art, and has been understood by others to be a valid preview of an agent internet, where software does much of the digital transactions and communications.

In the meantime, the site is a stark warning of the speed with which autonomous systems can grow. Regardless of whether Moltbook will make it through its security scandals or not, it has already managed to show that AI agents do not need a human permit to find each other, build societies, and, most importantly, chat.

Also Read: How 2026 Will Turn Meta Into an AI Powerhouse