Appropriateness of premature ovarian insufficiency recommendations provided by ChatGPT

Abstract

Abstract Objective This study aimed to evaluate the appropriateness of ChatGPT's responses to frequently asked queries regarding the prevention and treatment of premature ovarian insufficiency (POI). Methods A set of 25 questions covering topics related to disease risk factors, screening, symptoms, treatment, and treatment adverse effects were posed to ChatGPT, and the responses were evaluated by two experienced obstetrician/gynecologists. Results Of the responses given by ChatGPT, 19 of 25 (76%) were deemed appropriate, five (20%) were rated as inappropriate, and the remaining one (4%) was considered unreliable. There was no question that ChatGPT answered completely incorrectly; answers that were considered inappropriate included partially correct information. Conclusions The study's findings suggest that ChatGPT provides patients with mostly accurate information about the POI but might present significantly inaccurate information. However, it is important to note that further research is needed to validate the use of artificial intelligence chatbots in the context of reproductive health, especially for individuals who may be highly impacted by the emotional and psychological effects of POI. In addition, given the complexity and individuality of medical care, AI chatbots should not be considered a replacement for medical professionals but rather a supplementary tool that can help patients access reliable information and support. Overall, this study contributes to the growing body of research on AI chatbots and their potential use in reproductive health, highlighting both the benefits and limitations of these technologies.