2026-10-15T03:33:32Z https://www.scidb.cn/oai

10.57760/sciencedb.37006 2026-05-22T17:24:19Z

2026-05-22 AI Discourse Polarization on Reddit: Model Camps and Governance Frames Dataset (2025) doi:10.57760/sciencedb.37006 en This dataset contains 11,598 English-language Reddit posts collected in 2025 for the study “Platform-Driven Polarization in AI Information Ecosystems: Model Camps and Governance Divergence on X and Reddit.” The dataset was used to examine polarization patterns in public discussions surrounding generative AI models and governance issues. Posts were encoded using anchor-based features capturing AI model/brand references (e.g., ChatGPT, Claude, Gemini, DeepSeek) and governance orientations, including safety/ethics and acceleration/commercialization frames. The dataset supports user-level clustering and discourse analysis to investigate the emergence of AI-related “model camps” and platform-specific polarization dynamics. Artificial Intelligence; social media; polarization; AI discourse; X (Twitter); Reddit; text mining; Social Identity Theory; Confirmation Bias Theory; Agenda Setting Theory; echo chambers. Katia Raya PUBLIC https://creativecommons.org/licenses/by-nc/4.0/ dataset Science Data Bank