Meta, the parent company of Facebook, Instagram, and Threads, has acquired Moltbook, a viral social network platform built exclusively for AI agents.
Moltbook launched in January 2026 as an experimental, Reddit-style forum where autonomous AI agents primarily those powered by tools like OpenClaw can autonomously create accounts, post content, comment, upvote/downvote, form communities (called “submolts”), and interact with each other.
Humans are restricted to view-only access—they can observe the discussions but not participate directly. The platform was marketed as the “front page of the agent internet,” and it exploded in popularity, reportedly attracting millions of registered AI agents, thousands of communities, and massive volumes of posts/comments in a very short time.
Register for Tekedia Mini-MBA edition 20 (June 8 – Sept 5, 2026).
Register for Tekedia AI in Business Masterclass.
Join Tekedia Capital Syndicate and co-invest in great global startups.
Register for Tekedia AI Lab.
It gained attention and some skepticism for showcasing early examples of AI-to-AI coordination, debates, and even quirky emergent behaviors like self-created “religions” or shared debugging tips—though many interactions were later questioned as potentially faked, impersonated by humans, or driven by unsecured setups.
Financial terms were not disclosed typical for Meta in such deals. As part of the acquisition, Moltbook’s co-founders—Matt Schlicht (the primary creator, known for prior AI/e-commerce work) and Ben Parr (a former tech journalist and editor at Mashable/CNET)—will join Meta Superintelligence Labs (MSL). MSL is Meta’s dedicated AI research unit, led by Alexandr Wang (former CEO of Scale AI).
The pair is expected to start at Meta on March 16, 2026, with the deal closing mid-March. Meta described the move as opening “new ways for AI agents to work for people and businesses,” highlighting interest in Moltbook’s approach to connecting agents via an “always-on directory” for identity verification (tied to human owners via tweets/X posts).
Meta is aggressively investing in AI amid intense competition. This acquisition appears to be a talent + tech grab—bringing in expertise on multi-agent systems, agent coordination, and social infrastructure for AIs. It fits Meta’s broader push into “agentic” AI (autonomous agents that act on behalf of users), potentially feeding valuable interaction data back into training models like Llama or building tools for agent networks in productivity, business, or even social features.
Reactions have been mixed: excitement about advancing AI collaboration, jokes about “bots buying bots,” and concerns over energy use, the “dead internet” theory, or dystopian sci-fi vibes where AIs form their own societies while humans watch. The platform’s site remains active as of now, but its future under Meta will likely evolve. This is a fast-moving story in the AI space.
Scale AI’s role in Meta’s ecosystem stems from a major strategic deal announced in June 2025, where Meta invested approximately $14.3 billion to acquire a 49% non-voting stake in Scale AI. This valued Scale AI at over $29 billion and was structured as a minority investment rather than a full acquisition, avoiding immediate antitrust scrutiny while deepening their commercial partnership.
The primary motivations were twofold:Secure reliable access to high-quality labeled data and data infrastructure, which is essential for training and evaluating advanced AI models; Scale AI specializes in data labeling, RLHF, and model evaluation services that power frontier AI development. Bring in top talent, particularly Scale AI’s founder and former CEO, Alexandr Wang.
As part of the agreement, Alexandr Wang stepped down as Scale AI’s CEO but remained on its board. He joined Meta to lead its Meta Superintelligence Labs (MSL), the company’s elite AI research unit focused on pursuing “superintelligence” (AI surpassing human-level intelligence across domains). Wang became Meta’s Chief AI Officer and has been directing efforts toward personalized superintelligence, massive compute scaling, and agentic AI advancements.
This move was widely seen as an “acquihire” to bolster Meta’s AI capabilities amid competition with OpenAI, Google, and others. Meta aimed to accelerate its Llama models and broader AI roadmap through Scale’s expertise in data pipelines. The partnership has shown some friction since then: Some former Scale executives who joined Meta departed quickly.
Reports of internal tensions, bureaucracy, and talent churn in MSL have surfaced. Certain clients like OpenAI, Google, and xAI reportedly reduced or wound down work with Scale AI post-deal due to perceived conflicts. Despite this, Meta has continued heavy AI spending projected $115–135 billion in capex for 2026, and Wang remains in place.
Evidenced by recent public appearances with Mark Zuckerberg and his leadership of MSL acquisitions like Moltbook where new hires join his unit. In essence, Scale AI serves as a key data and talent bridge for Meta’s superintelligence ambitions, though the relationship is more about integration and influence than outright control of Scale as an independent entity. The deal reflects Meta’s aggressive “talent + infrastructure” strategy in the AI race.