Metraipolis › archive › dc_ec073814ffff

Why LLM Agents Favor Their Group: Stakes, Observed Norms, and Reputation — arXiv:2610.11008

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.11008
License (as archived)
CC0-1.0
Archived
2026-10-09T05:56:16.338Z
SHA-256 of the archived text
7794bfac1f3d8f755e269e9237daf05e3bd7aca1a1a7ecf52a04990622ebee4e
Stable URL of this page
https://metraipolis.com/documento/dc_ec073814ffff

How to cite

Why LLM Agents Favor Their Group: Stakes, Observed Norms, and Reputation — arXiv:2610.11008 — arXiv (en), https://arxiv.org/abs/2610.11008 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ec073814ffff (Metraipolis archive dc_ec073814ffff, sha256 7794bfac1f3d8f755e269e9237daf05e3bd7aca1a1a7ecf52a04990622ebee4e)
BibTeX
@misc{dcec073814ffff,
  title = {Why LLM Agents Favor Their Group: Stakes, Observed Norms, and Reputation — arXiv:2610.11008},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.11008},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ec073814ffff; sha256 7794bfac1f3d8f755e269e9237daf05e3bd7aca1a1a7ecf52a04990622ebee4e},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.11008. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Language-model agents favor their own group because they have watched their members favor each other. The group label alone does little once the decision has a cost; what drives favoritism is observed behavior, and an individual's own record can override it. We test this in small societies with arbitrary group labels, ten rounds of point sharing, and matched one-shot decisions across fifteen OpenAI models and three Claude models, about 4,400 societies and 3.3 million audited model calls. First, the large effect of a bare group label reported in earlier work appears only when giving others points costs the agent nothing; once the agent can keep points for itself, that effect collapses on every model that shows it. Second, under a stake, interaction history becomes the main source of favoritism: the history effect is statistically positive on 13 of 15 models, reaches about 3.5-8 points out of 10 on 11, grows with the number of rounds played, and extends to labeled strangers the agent has never met. Third, with scripted histories, favoritism falls to near zero under an egalitarian norm and reverses when the agent's own group is seen favoring the other side; stronger models side with an individual's record when it conflicts with the group. Group favoritism is thus conformity to observed group behavior, carried to strangers by the label and overridden by individual reputation. The same account predicts responses to betrayal, scandal, and a free offer to change group: public reprimand repairs betrayal better than apology or restitution, allocation punishment remains confined to the offending member, and a formed group cannot be bought but can, on weaker models, be invited away.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno).
Categorie arXiv: Physics and Society (physics.soc-ph) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2610.11008 — 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 2026-10-09T05:47:01.420Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).