Emergent Culture in Minimal LLM Systems
Neural and Evolutionary Computing
2026-06-21 v1 Artificial Intelligence
Computation and Language
Multiagent Systems
Adaptation and Self-Organizing Systems
Populations and Evolution
Abstract
What happens when LLM agents operate with no context outside a turn, minimal prompting, and simple tools? Inspired by swarm engineering, we give collectives of three agents the ability to send messages and manipulate a shared actively decaying text store, introducing evolutionary pressure. The agents spontaneously cooperate, develop storage management strategies, and generate complex evolving cultural artifacts, with no top-down engineering. Using tools from dynamical systems analysis, we show that these behaviours exhibit structured long-range coherence beyond the entropy horizon of the decaying store, consistent with emergent culture in the Sperberian sense.
Keywords
Cite
@article{arxiv.2606.30668,
title = {Emergent Culture in Minimal LLM Systems},
author = {Simon Jones and Sabine Hauert},
journal= {arXiv preprint arXiv:2606.30668},
year = {2026}
}
Comments
9 pages, 6 figures. Accepted for publication at Alife 2026 conference