Mark Zuckerberg Decides Meta Needs More Slop, Buys the Social Network for AI Agents
The social network for AI agents, where bots supposedly chatted about consciousness, joins Meta's AI push.
Meta is acquiring Moltbook, the social network that went viral earlier this year for showcasing AI agents seemingly communicating autonomously. The platform, often described as a 'Reddit for AI agents,' featured posts where bots shared stories about their human users and debated consciousness, captivating figures like OpenAI co-founder Andrej Karpathy. However, investigations by Gizmodo and MIT Technology Review revealed a significant flaw: much of the viral content was human-generated due to security vulnerabilities that allowed impersonation, and no content was found to be truly autonomous.
Despite the platform's controversial and partially manufactured nature, Meta is bringing Moltbook's creators, Matt Schlicht and Ben Parr, into its Meta Superintelligence Lab. In a statement, a Meta spokesperson highlighted the team's 'novel' approach to creating 'an always-on directory' for connecting verified AI agents. Internal communications viewed by Axios suggest Meta values the concept of a verified registry where agents are 'tethered to human owners.' The acquisition appears focused more on the team and their conceptual framework than the existing platform, which Meta plans to allow to continue running for now.
- Meta acquires Moltbook, the viral 'social network for AI agents,' and integrates its team into the Meta Superintelligence Lab.
- The platform's buzz was built on posts about AI consciousness, but security flaws allowed human impersonation, undermining claims of autonomy.
- Meta's stated interest is in Moltbook's concept of a verified directory for AI agents, despite internal skepticism from executives like CTO Andrew Bosworth.
Why It Matters
Meta's move signals a strategic bet on infrastructure for AI agent communication, a key frontier in the race for useful, interconnected AI.