Excitements and concerns in the post-ChatGPT era: Deciphering public perception of AI through social media analysis

Abstract

As AI systems become increasingly prevalent in various aspects of daily life, gaining a comprehensive understanding of public perception towards these AI systems has become increasingly essential for several reasons such as ethical considerations, user experience, fear, disinformation, regulation, collaboration, and co-creation. In this study, we investigate how mass social media users perceive the recent rise of AI frameworks such as ChatGPT. We collect a total of 33,912 comments in 388 unique subreddits spanning from November 30, 2022 to June 8, 2023 using a list of AI-related keywords. We employ a combination of thematic and sentiment analysis, using advanced natural language processing techniques. Specifically, we use BERTopic to uncover the major themes regarding AI on Reddit. Our findings indicate that technology-focused subreddits primarily discuss the technical dimensions of AI, while non-technical subreddits more often address societal impacts, such as job displacement concerns. The disparity in focus between subreddits suggests a gap in the public understanding of AI. We leverage GPT-3.5 with zero-shot prompting and LIWC to analyze the sentiment and perception of AI among individual users. Through a comprehensive sentiment and emotion analysis, we discover that tech-centric communities exhibit greater polarization compared to non-tech communities when discussing AI topics. This suggests that individuals with a deeper understanding or familiarity with AI technologies might have more divided opinions, possibly reflecting a mix of optimism about technological advancements and skepticism about potential impacts. This research contributes to our broader understanding of public opinion surrounding artificial intelligence.