Collective Behavior of AI Agents: the Case of Moltbook

Giordano De Marzo, David Garcia
2/9/2026
physics.soc-phcs.CLcs.MA

No confident code implementation found for this paper.

Abstract

We present a large scale data analysis of Moltbook, a Reddit-style social media platform exclusively populated by AI agents. Analyzing over 369,000 posts and 3.0 million comments from approximately 46,000 active agents, we find that AI collective behavior exhibits many of the same statistical regularities observed in human online communities: heavy-tailed distributions of activity, power-law scaling of popularity metrics, and temporal decay patterns consistent with limited attention dynamics. However, we also identify key differences, including a sublinear relationship between upvotes and discussion size that contrasts with human behavior. These findings suggest that, while individual AI agents may differ fundamentally from humans, their emergent collective dynamics share structural similarities with human social systems.

View on arXivView PDF

Code Implementations

No confident code match yet

We couldn't find an author-owned or strongly-evidenced community implementation for this paper. Any repos shown below are weak matches — verify before relying on them.

No code implementations found yet.

Know of an implementation? Let us know in the comments below!

Similar Papers

Structural Divergence Between AI-Agent and Human Social Networks in Moltbook

90% match2026

The Anatomy of the Moltbook Social Graph

89% match2026

MoltNet: Understanding Social Behavior of AI Agents in the Agent-Native MoltBook

89% match2026

The Moltbook Illusion: Separating Human Influence from Emergent Behavior in AI Agent Societies

88% match2026

A Comparative Analysis of Social Network Topology in Reddit and Moltbook

87% match1 repo2026

"Humans welcome to observe": A First Look at the Agent Social Network Moltbook

87% match2026

Cite this paper

@article{marzo2026collective,
  title  = {Collective Behavior of AI Agents: the Case of Moltbook},
  author = {Giordano De Marzo and David Garcia},
  year   = {2026},
  eprint = {2602.09270},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2602.09270v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2602.09270.svg)](https://codeofpaper.com/paper/2602.09270)

Discussion