{"paper":{"title":"How Large Language Models play humans in online conversations: a simulated study of the 2016 US politics on Reddit","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CY","cs.SI","physics.soc-ph"],"primary_cat":"cs.CL","authors_text":"Daniele Cirulli, Giovanni Palermo, Giulio Cimini","submitted_at":"2025-06-23T08:54:32Z","abstract_excerpt":"Large Language Models (LLMs) have recently emerged as powerful tools for natural language generation, with applications spanning from content creation to social simulations. Their ability to mimic human interactions raises both opportunities and concerns, particularly in the context of politically relevant online discussions. In this study, we evaluate the performance of LLMs in replicating user-generated content within a real-world, divisive scenario: Reddit conversations during the 2016 US Presidential election. In particular, we conduct three different experiments, asking GPT-4 to generate "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.21620","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2506.21620/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}