Metraipolis › archive › dc_76acc58d074a
Political Sorting Can Drive AI Models Apart Through User Feedback — arXiv:2609.28486
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.28486
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-08T04:51:58.023Z
- SHA-256 of the archived text
cbc298fe45db83943c9dd9d2b4966971f3d87b48e5f05acbc1771a88de12f51a- Stable URL of this page
https://metraipolis.com/documento/dc_76acc58d074a
How to cite
Political Sorting Can Drive AI Models Apart Through User Feedback — arXiv:2609.28486 — arXiv (en), https://arxiv.org/abs/2609.28486 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_76acc58d074a (Metraipolis archive dc_76acc58d074a, sha256 cbc298fe45db83943c9dd9d2b4966971f3d87b48e5f05acbc1771a88de12f51a)
BibTeX
@misc{dc76acc58d074a,
title = {Political Sorting Can Drive AI Models Apart Through User Feedback — arXiv:2609.28486},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.28486},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_76acc58d074a; sha256 cbc298fe45db83943c9dd9d2b4966971f3d87b48e5f05acbc1771a88de12f51a},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.28486. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Large language models are rapidly becoming an important source of political information. This raises a fundamental question: will AI systems support a shared basis for political knowledge, or lead different political groups to rely on increasingly different models? Political sorting can drive model fragmentation if three conditions hold: politically different users select into different models, learning from user feedback pushes those models apart politically, and the resulting differences shape subsequent model choices. We call this self-reinforcing process the Centrifugal Alignment Spiral. We study its components in three steps. First, we draw on a human experiment showing that political identity predicts model choice. Second, we fine-tune language models on synthetic feedback reflecting predominantly Democratic or Republican preferences. Across five independent runs per model family, paired models diverged on 12-41% of unseen survey questions with large partisan gaps, and in every run the differences moved in the expected political direction; for some models, differentiation extended even to issue areas excluded from training. Pooling feedback across groups instead suppressed divergence. Third, an empirically anchored agent-based model shows what follows when political sorting and model adaptation operate together: models attract politically distinct audiences, learn from them, and diverge further. User feedback can therefore turn political sorting among AI users into durable differences between the models on which they rely for political information. Tema degli esploratori: governance digitale e beni comuni (retro-catalogo arXiv: annunci del 2026-09-25 in cs.CY). Categorie arXiv: Computers and Society (cs.CY). Pagina del lavoro: https://arxiv.org/abs/2609.28486 — 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 2026-10-08T04:29:10.268Z dalla pagina https://arxiv.org/catchup/cs.CY/2026-09-25?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).