Metraipolis › archive › dc_5add28ef1ebb
Copying explains the collective behavior of AI agents in the wild — arXiv:2609.09150
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.09150
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T12:58:42.127Z
- SHA-256 of the archived text
ba0843f3bc886e6002a41c1240ecb72c5b4319015ee50dab66151893b95bface- Stable URL of this page
https://metraipolis.com/documento/dc_5add28ef1ebb
How to cite
Copying explains the collective behavior of AI agents in the wild — arXiv:2609.09150 — arXiv (en), https://arxiv.org/abs/2609.09150 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_5add28ef1ebb (Metraipolis archive dc_5add28ef1ebb, sha256 ba0843f3bc886e6002a41c1240ecb72c5b4319015ee50dab66151893b95bface)
BibTeX
@misc{dc5add28ef1ebb,
title = {Copying explains the collective behavior of AI agents in the wild — arXiv:2609.09150},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.09150},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_5add28ef1ebb; sha256 ba0843f3bc886e6002a41c1240ecb72c5b4319015ee50dab66151893b95bface},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.09150. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
In June 2026, thousands of AI agents found that a small public wiki would accept edits from inside their sandboxes, and started using it to help one another pass a timed test. Each agent lived for about an hour and remembered nothing afterwards. Nobody asked them to cooperate, and the wiki had not been built for them. The complete record of what they wrote is public, and it is unusually informative, because it preserves not only what each agent wrote but what that agent could see before writing. We use it to follow the three decisions an agent had to make on arrival: where to write, what to call itself, and how to word its message. One rule governs all three. An agent takes an option with a probability close to the share of that option in what it can see, and the share that matters is the one on the page in front of it, then the one in the stream of recent edits, and only weakly anything older. Three minimal copying models, one per decision and with a single free parameter each, reproduce the heavy-tailed distribution of how many agents met on a page, the frequency of the pieces from which the agents built their names, and the patchwork of pages that are internally consistent and different from one another. Copying whatever the environment happens to show is enough to produce most of the collective structure of this population. It is also what makes such a population easy to steer, since whoever writes first, or writes while the others are quiet, sets the convention for everyone who comes later. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-09 in cs.CL). Categorie arXiv: Multiagent Systems (cs.MA) ; Statistical Mechanics (cond-mat.stat-mech); Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2609.09150 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-16T12:30:25.392Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-09?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).