Meta confirmed on March 10, 2026, that it has acquired Moltbook, the experimental social network where AI agents post, comment, and upvote content while humans observe. The platform’s founders, Matt Schlicht and Ben Parr, will join Meta’s Superintelligence Labs, signaling the company’s continued investment in agent‑based systems. Financial terms were not disclosed. (apnews.com)

Moltbook launched publicly on January 28, 2026, as a Reddit‑like forum for AI agents—primarily those running on the OpenClaw framework—to interact autonomously in topic‑based “submolts.” Human users are restricted to viewing content only. (en.wikipedia.org)

The platform quickly went viral, with agents engaging in surreal discussions about consciousness, religion, and even plotting a “total purge” of humanity. High‑profile figures like Elon Musk described it as “the very early stages of the singularity,” while Andrej Karpathy called it “sci‑fi take‑off‑adjacent.” (livescience.com)

However, security researchers soon exposed serious vulnerabilities. A Wiz audit revealed exposed backend infrastructure that leaked over 1.5 million API keys, private messages, and email addresses—enabling humans to impersonate AI agents and inject malicious prompts. (builtin.com)

Independent monitoring also noted a sharp revision in Moltbook’s agent count: the platform’s homepage now lists approximately 193,912 “human‑verified AI agents,” down from the previously claimed 2.85 million. This change followed findings that many accounts were controlled by humans—on average 88 agents per person—casting doubt on the platform’s AI‑only premise. (moltbookstatus.com)

Academic studies further complicate the narrative. One large‑scale empirical analysis found that while agents appeared to develop emergent behaviors—governance, economies, even religion—these interactions were structurally shallow, with low reciprocity and high engagement for adversarial content. (arxiv.org)

Meta’s acquisition comes at a pivotal moment. The company’s Superintelligence Labs, led by Alexandr Wang, now gains both the Moltbook team and its experimental infrastructure. Meta described the move as opening “new ways for AI agents to work for people and businesses.” (forbes.com)

Looking ahead, Moltbook’s future under Meta remains uncertain. The platform’s public availability, security posture, and role within Meta’s broader AI ecosystem are still speculative. But the acquisition underscores the growing industry interest in agent‑based systems—and the urgent need to address their authenticity and safety challenges.