Response data for LLMs left, right, and center: Assessing GPT's capabilities to label political bias from web domains

This research investigates whether OpenAI's GPT-4, a state-of-the-art large language model, can accurately classify the political bias of news sources based solely on their URLs. This dataset contains GPT-4 classifications for sources listed on Media Bias/Fact Check (MBFC), rated on a seven-degree scale ("far-left" to "far-right").

本研究旨在探究OpenAI推出的GPT-4——当前最先进的大语言模型（Large Language Model）——能否仅凭借新闻来源的统一资源定位符（Uniform Resource Locator，URL），准确判定其政治倾向。本数据集包含媒体偏见与事实核查（Media Bias/Fact Check，MBFC）平台收录的全部新闻来源的GPT-4分类结果，所有分类均采用七级评分量表，评级区间覆盖“极左翼”至“极右翼”。