Despite recent advances in understanding the capabilities and limits of generative artificial intelligence (GenAI) models, we are just beginning to understand their capacity to assess and reason about the veracity of content. We evaluate multiple GenAI models across tasks that involve the rating of, and reasoning about, the credibility of information. The information in our experiments comes from content that subnational U.S. politicians post to Facebook. We find that GPT-4o outperforms other models, but all models exhibit only moderate agreement with human coders.
The rise in hate speech targeting minority communities underscores the urgent need for effective tools to detect and address harmful content in digital communication. We examine over 3 million tweets posted by state legislators between 2020 and 2021, focusing on messages directed at Asian communities. To address the nuanced nature of hate speech, we develop three comprehensive definitions for identifying hate speech. With a human-in-the-loop approach, our fine-tuned BERT-NLI model achieved improved classification performance.
Amid national gridlock on climate policy, state legislatures have become increasingly central to climate governance—yet we know little about how state legislators communicate climate issues online. Drawing on 6,177,988 social media posts from 6,353 legislators on Facebook and X (formerly Twitter) between 2020 and 2023, we fine-tune a large language model to classify climate-relevant content and stance. Among nearly 58,000 labeled posts between 2020 and 2021, 34,165 were supportive and 11,573 opposing, with clear variation across platforms.
Recent work in digital politics has begun to explore the role of race and ethnicity in digital communications. This research, however, has not fully addressed how lawmakers interact with their Asian constituents and the broader minority population. We take up this task by analyzing over 3 million tweets posted by state legislators between 2020 and 2021, focusing on messages targeting Asian ethnic groups. We fine-tuned a large language on classification task by using our labelled data and detected 7,202 anti-racism speech and 2,536 racism speech among 25,102 tweets that target Asian ethnic groups specifically.
This project conducts a literature review of Large Language Models (LLMs) to explore how Generative Artificial Intelligence (AI) has been discussed across sub disciplines of social science, specifically in Communication and political science. A pertinent topic, the gap in literature reviews is addressed in this study. Furthermore, the comparison provides a baseline of examination of key topics, differences, and the role of ethics and GenAI between the disciplines. The irony of utilizing computational methods in this paper is not lost and provides robust methods of empirical examination.