By Interestana AI Editorial — AI-drafted, human-overseen. How we report
Fox News Attacks Media Every 14.5 Minutes, Study Finds

Fox News engages in inter-journalism disparagement, or attacks against other news outlets and the media in general, at a rate of approximately once every 14.5 minutes, according to a recent study by Duke University’s DeWitt Wallace Center for Media & Democracy. This phenomenon, identified as a potential driver of declining trust in journalism, was analyzed across CNN, Fox News, and MSNBC/MS NOW using 15 years of cable news transcripts from 2011 to 2025. The research team collected a random sample of one week's programming for each year, resulting in 3,543 transcripts and nearly 19 million words, equivalent to approximately 62,000 pages of text. To process this extensive data, the researchers trained OpenAI’s GPT-5.5 model to identify instances of disparagement, developing a 20-page instruction document to guide the AI. GPT-5.5 achieved a 99.5% accuracy rate in identifying inter-journalism disparagement and a 100% accuracy rate in avoiding false negatives when compared to human-coded samples. In contrast to Fox News, CNN was found to engage in similar disparagement approximately every 140 minutes, while MSNBC did so roughly every 61 minutes. The study highlights a significant disparity in how these major cable news networks discuss the media landscape. The findings emerge at a time when trust in journalism has reached historically low levels. A Gallup poll from 2025 indicated that only 8% of Republicans expressed trust in the news media, compared to 51% of Democrats. While recent data show a slight uptick among Republicans, overall trust remains near this record low, and the partisan gap persists. The Duke University study aimed to quantify the extent of inter-journalism disparagement and its variation across networks, contributing to a deeper understanding of the factors influencing public perception of news organizations. The methodology involved a comprehensive analysis of broadcast content, leveraging advanced AI for large-scale text processing and classification. The researchers defined inter-journalism disparagement to encompass direct criticisms, negative framing, and dismissive commentary directed at other news entities or the profession of journalism as a whole. The study's detailed analytical process, including the AI training and validation, is available for further review. This research provides concrete data on media self-critique and its potential impact on journalistic credibility in an era of polarized media consumption.
Original source — read the full reporting at the publisher:
Read on Nieman LabGet the weekly AI digest
AI news + new model releases, weekly. Drafted by our agents, reviewed by humans.