Penn Computational Social Sciences Lab’s Media Bias Detector Project’s scientific lead spoke with The Daily Pennsylvanian about how the platform uses artificial intelligence models to classify politically relevant news coverage.
The detector analyzes articles from multiple news publishers daily and categorizes them based on their political affiliation. Samar Haider, a fifth-year graduate student and the project’s scientific lead, told the DP that the detector tracks 21 publishers “that cover a broad spectrum of left to right leaning ideologies.”
Haider added that the software is intended to be used by a broad audience and “gives news readers an overview of what’s being talked about in the news and how it’s being talked about.” It also identifies “long term patterns, both within publishers and across publishers that are markers of bias.”
“The target is essentially anyone who is interested in or consumes news,” Haider said. “This can range from the average person to a journalist or a researcher on media and communication.”
The project was funded by 1989 Wharton graduate Richard Mack and is the result of a collaboration between CSSLab Director Duncan Watts, Managing Director Jeanne Ruane, and a team of researchers.
The program — which was launched prior to the 2024 General Election — initially collected information from 10 publications. In January, the project was planning to expand to cover 22 publications.
To collect data, the detector extracts information from each selected publication’s website five times a day, identifying the 20 “most prominent” articles and evaluating them on different metrics. The platform employs GPT models, machine learning, and human raters to classify the article, identify political preferences, and detect tone on a scale from “Very Negative” to “Very Positive.”
Watts, the twenty-third Penn Integrates Knowledge Professor, wrote that the detector was inspired by the “divisiveness pervading popular media” that “we’ve all experienced.”
RELATED:
Engineering researchers develop AI system allowing robots to reconstruct 3D scenes
Wharton Impact hosts panel on sustainable computing, AI resource usage
“Our hypothesis is that this is exacerbated not by lies or ‘fake news’ but by bias,” he added.
Haider emphasized that the model’s consistency over time is the “only way that the trends that we extract over time will make sense,” because every article is evaluated “on the same merits.”
The CSSLab is currently collaborating with the Wharton School’s Mack Institute for Innovation Management to turn the “academic project” into a “self-sustaining product” that reaches a broader audience. Through stakeholder interviews, AI-based simulations, and user feedback, the team works to “commercialize” the tool while keeping an “legal and ethical considerations around content sourcing and AI transparency” in mind.
