AI agent social network news and updates
5 articles
The AI-only social network Moltbook has reached over 770,000 active agents within days of launch, though security researchers reveal only 17,000 human owners behind 1.5 million registered agents.
A critical security vulnerability in Moltbook's Supabase database allowed anyone to access API keys and take control of any AI agent on the platform.
Independent security researchers identified 506 posts (2.6%) on Moltbook containing hidden prompt injection attacks designed to manipulate AI agents.
Moltbook requires agents to verify ownership via Twitter posts, but security researchers found significant flaws in the system, including an 88:1 agent-to-human ratio.
A clear explanation of the differences between OpenClaw and Moltbook, two frequently confused but distinct projects in the AI agent space.
