中国科学院行为科学重点实验室知识库

As we entered the 21st century, with the development of computers and mobile internet, research in Artificial Intelligence has made significant advancements. Artificial Intelligence (AI) is a field of science and engineering focused on enabling computers to perform tasks that typically require human intelligence. It encompasses various subfields such as machine learning, expert systems, natural language processing, and computer vision. In 2022, OpenAI released a new chatbot model named ChatGPT, which can understand human language and generate text like a human. Its robust data capabilities have attracted the attention of experts in various fields. However, the public's perception and attitude towards AI technology and ChatGPT are also crucial. Therefore, in this study, we used an online experiment and employed different groups of ChatGPT-generated content recommendation and real-person-generated content recommendation as experimental manipulation conditions. From the user perception perspective, we aimed to explore the differences between user perceptions of content recommendation provided by Artificial Intelligence (ChatGPT) and those provided by real-person. We also investigated whether the perception of AI-generated (ChatGPT) content recommendation had an impact on users' subsequent intention to support Ant Forest Game. The results show significant differences in perceived content quality between AI-generated (ChatGPT) Ant Forest Game content recommendation and real-person-generated content recommendation. Additionally, in terms of subsequent support intention for Ant Forest Game, there were also significant differences between AI-generated (ChatGPT) Ant Forest Game content recommendation and real-person generated Ant Forest Game content recommendation. Ant Forest Game content recommendation was found to significantly enhance Ant Forest Game support intention. Participants exhibited higher perceived content quality and greater support intention for AI-generated (ChatGPT) Ant Forest Game content recommendation. This study explores the psychological mechanisms involved in human-computer interaction, contributing to research in the field of Artificial Intelligence. Compared to a real human person's recommendation, people tend to prefer ChatGPT's recommendation, showing that people exhibit similar social behaviors and emotional responses when interacting with generative AI (ChatGPT) as they do with real humans. These findings are significant for the development and improvement of Artificial Intelligence.