FakeGPT: Fake News Generation, Explanation and Detection of Large Language Models
Yue Huang, Lichao Sun
TL;DR
This study probes ChatGPT's dual-use potential in fake news by examining generation, explanation, and detection. It demonstrates that carefully crafted prompts can produce highly realistic fake news, identifies nine explainable reasons behind fake content, and introduces a reason-aware prompting approach that improves detection across multiple datasets, with supplementary contextual information providing additional gains. The work highlights both the misuse risk and the opportunity to leverage LLMs for disinformation analysis, while acknowledging limitations such as dataset scope and model transparency. Overall, it offers actionable insights into prompt design and information augmentation to enhance fake-news detection, and calls for responsible governance and further research.
Abstract
The rampant spread of fake news has adversely affected society, resulting in extensive research on curbing its spread. As a notable milestone in large language models (LLMs), ChatGPT has gained significant attention due to its exceptional natural language processing capabilities. In this study, we present a thorough exploration of ChatGPT's proficiency in generating, explaining, and detecting fake news as follows. Generation -- We employ four prompt methods to generate fake news samples and prove the high quality of these samples through both self-assessment and human evaluation. Explanation -- We obtain nine features to characterize fake news based on ChatGPT's explanations and analyze the distribution of these factors across multiple public datasets. Detection -- We examine ChatGPT's capacity to identify fake news. We explore its detection consistency and then propose a reason-aware prompt method to improve its performance. Although our experiments demonstrate that ChatGPT shows commendable performance in detecting fake news, there is still room for its improvement. Consequently, we further probe into the potential extra information that could bolster its effectiveness in detecting fake news.