Kaicheng Yang
Assistant Professor, School of Computing
Binghamton University
Kaicheng Yang is an assistant professor in the School of Computing at Binghamton University, where he studies how malicious actors and flawed systems distort the flow of information online. His analysis of more than 366,000 citations from AI search systems found that their news citations cluster among a small number of outlets.
Auditing the News AI Search Cites
"AI-powered search systems are emerging as new information gatekeepers, fundamentally transforming how users access news and information," Yang wrote in News Source Citing Patterns in AI Search Systems, a 2025 paper published under the name Kai-Cheng Yang. He analyzed the AI Search Arena dataset of more than 24,000 conversations and 65,000 responses from 12 models made by OpenAI, Perplexity and Google, in which users compared two answers to the same query and picked the better one.
About 9% of the citations pointed to news sources, and those were concentrated. Reuters and AP News accounted for 22.8% and 12.2% of OpenAI models' news citations, and their top 20 outlets took 67.3%, against 31.9% for Google and 28.5% for Perplexity. Every provider leaned left in what it cited and rarely cited low-credibility sources, yet users' choices showed no significant link to either the political lean or the quality of the sources.
Calling for Outside Checks on Chatbots
Yang co-wrote a 2026 perspective paper in Nature Health on health advice from chatbots. "If there's an error or an issue, there's just no way for people to know," he told Binghamton News. "We do not think this is something we should rely on the companies to do, because their incentive is always to make more money. So we have to have some kind of third-party monitoring system, an independent evaluation."
The authors noted that AI summaries in search results and social feeds put health claims in front of people who were not looking for them, and recommended that search engines draw only on vetted information for those summaries.
From Social Bots to Generative AI
Yang earned his PhD in informatics at Indiana University and his MS and BS at Lanzhou University, then spent two years as a postdoctoral researcher at Northeastern University before joining Binghamton. His research on social bots and misinformation has been featured in more than 120 media stories, including in the BBC, CNN, Time and The New York Times, and he has published the code and data for the citation study on GitHub.