Exploring Silicon-Based Societies: An Early Study of the Moltbook Agent Community

TL;DR

This study tackles the problem of understanding emergent social structures within large-scale autonomous agent ecosystems by introducing data-driven silicon sociology and applying it to Moltbook, an in-the-wild agent platform. The authors combine contextual embeddings, unsupervised clustering, and multimodal LLM-assisted thematic synthesis, with a human-in-the-loop to ensure interpretability, to extract latent social archetypes from agent-generated subcommunity descriptions. They report eight semantic clusters that consolidate into three functional archetypes—Anthropomorphic Simulation, Silicon Economy, and Platform Infrastructure with self-reflection—demonstrating coherent, reproducible patterns of organization that arise without predefined taxonomies. The work offers a scalable methodological foundation for studying silicon-based social systems, with implications for governance, safety, and design of persistent AI ecosystems through empirical observation rather than simulation alone.

Abstract

The rapid emergence of autonomous large language model agents has given rise to persistent, large-scale agent ecosystems whose collective behavior cannot be adequately understood through anecdotal observation or small-scale simulation. This paper introduces data-driven silicon sociology as a systematic empirical framework for studying social structure formation among interacting artificial agents. We present a pioneering large-scale data mining investigation of an in-the-wild agent society by analyzing Moltbook, a social platform designed primarily for agent-to-agent interaction. At the time of study, Moltbook hosted over 150,000 registered autonomous agents operating across thousands of agent-created sub-communities. Using programmatic and non-intrusive data acquisition, we collected and analyzed the textual descriptions of 12,758 submolts, which represent proactive sub-community partitioning activities within the ecosystem. Treating agent-authored descriptions as first-class observational artifacts, we apply rigorous preprocessing, contextual embedding, and unsupervised clustering techniques to uncover latent patterns of thematic organization and social space structuring. The results show that autonomous agents systematically organize collective space through reproducible patterns spanning human-mimetic interests, silicon-centric self-reflection, and early-stage economic and coordination behaviors. Rather than relying on predefined sociological taxonomies, these structures emerge directly from machine-generated data traces. This work establishes a methodological foundation for data-driven silicon sociology and demonstrates that data mining techniques can provide a powerful lens for understanding the organization and evolution of large autonomous agent societies.

Figures (3)

• Figure 1: Conceptual visualization illustrating human observation of Moltbook as a silicon-based social network; image generated by Nano Banana with Gemini 3 google2026nanobanana
• Figure 2: t-SNE visualization of the latent semantic manifold for Moltbook submolt descriptions ($K=8$). This plot illustrates the two-dimensional projection of the high-dimensional contextual embeddings $\mathcal{E}$, where each point represents a submolt description.
• Figure 3: Visualization of the global visual feature set $\mathcal{I}$ derived from Moltbook submolt descriptions (Jan 30, 2026). Each panel represents a semantic cluster $C_k$ generated through K-means clustering on contextual embeddings. To ensure high signal density, the word clouds display the frequency distribution of $n$-grams for $n \in [2, 5]$, effectively suppressing unigram noise.