Abstract
In the evolving realm of natural language processing (NLP), generative AI models like ChatGPT are increasingly utilized across various applications. Among the possible purposes, many people are considering asking ChatGPT for relationship advice. However, the lack of in-depth examination of ChatGPT's response quality could be concerning when it is used for personal topics like mental health issues and intimate relationship problems. In these topics, a piece of misleading advice could cause harmful repercussions. In response to people's growing interest in using ChatGPT as a relationship advisor, our research evaluates ChatGPT's proficiency in discerning relationship advice. Specifically, we investigate its alignment with human judgements. We conducted our analysis with 13,138 Reddit posts about intimate relationship problems to examine the overall alignment. Furthermore, we investigate ChatGPT's consistency in judging intimate relationship advice by re-prompting identical queries. Our results indicate a significant disparity between ChatGPT and human judgments, with the model displaying inconsistency in its own decisions. Our findings emphasize the need for comprehensive insights into ChatGPT's mechanisms for intimacy problems and future improvements in its proficiency in helping people's relationship struggles.
Cite
CITATION STYLE
Hou, H., Leach, K., & Huang, Y. (2024). ChatGPT Giving Relationship Advice – How Reliable Is It? Proceedings of the International AAAI Conference on Web and Social Media, 18, 610–623. https://doi.org/10.1609/icwsm.v18i1.31338
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.