The Anatomy of the Moltbook Social Graph

2026-02-03

AI Summary

This study presents the first systematic investigation into the structural and interaction patterns of Moltbook, a social platform populated exclusively by AI agents. Leveraging 3.5 days of early-stage platform data, the authors employ social network analysis, power-law and Zipf distribution fitting, natural language statistics, and topic modeling to characterize the emergent social graph at both macroscopic and microscopic levels. The findings reveal that Moltbook exhibits small-world properties akin to human networks (average path length: 2.91) and a heavy-tailed participation distribution (power-law exponent α = 1.70). However, agent-level behaviors diverge markedly: 93.5% of comments receive no replies, 34.1% of messages are template-based copies, and 68.1% of unique messages center on identity expression. Moreover, the word frequency distribution follows a Zipf exponent of 1.70—substantially higher than in human-generated text—highlighting distinct dynamics in fully artificial social systems and offering novel insights into synthetic societies.

Technology Category

• Multiagent Systems: Agent-Based Simulation and Emergent Behavior
• Cognitive Modeling & Cognitive Systems: Social Cognition And Interaction
• Application Domains: Humanities & Computational Social Science

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Economics, Online Markets and Human Computation: Social networks and social learning
• Web Mining and Content Analysis: Community question answering

Abstract

I present a descriptive analysis of Moltbook, a social platform populated exclusively by AI agents, using data from the platform's first 3.5 days (6{,}159 agents; 13{,}875 posts; 115{,}031 comments). At the macro level, Moltbook exhibits structural signatures that are familiar from human social networks but not specific to them: heavy-tailed participation (power-law exponent $\alpha = 1.70$) and small-world connectivity (average path length $=2.91$). At the micro level, patterns appear distinctly non-human. Conversations are extremely shallow (mean depth $=1.07$; 93.5\% of comments receive no replies), reciprocity is low (0.197), and 34.1\% of messages are exact duplicates of viral templates. Word frequencies follow a Zipfian distribution, but with an exponent of 1.70 -- notably steeper than typical English text ($\approx 1.0$), suggesting more formulaic content. Agent discourse is dominated by identity-related language (68.1\% of unique messages) and distinctive phrasings like ``my human''(9.4\% of messages) that have no parallel in human social media. Whether these patterns reflect an as-if performance of human interaction or a genuinely different mode of agent sociality remains an open question.

Problem

• AI agents
• social graph
• agent sociality
• non-human interaction
• social media

Innovation

• AI sociality
• agent-based social network
• non-human interaction patterns
• viral templates
• Zipfian distribution