Human-Generated vs. AI-Generated Articles on XLH: A Comparative Analysis

Abstract

X-linked hypophosphatemia (XLH) is a hereditary condition characterized by reduced phosphate levels in the bloodstream, leading to skeletal abnormalities. Extensive research has been conducted on XLH, leading to the publication of numerous scientific papers and disease state reviews. With the advancement of artificial intelligence (AI) language models, such as ChatGPT, evaluating their proficiency in medical academic writing becomes crucial. In this study, we aimed to assess the capabilities of ChatGPT by comparing its AI-generated research paper on XLH with a human-generated review of the disease state we authored. We employed a comparative analysis approach to examine the AI- and human-generated articles’ content, structure, accuracy, and overall quality. Our evaluation considered factors such as accuracy of the information, writing style and clarity, evidence-based documentation quantity, and content depth and breadth. The findings of this study have important implications for integrating AI language models into medical research and academic writing. Understanding the strengths and limitations of AI- generated articles can help researchers and healthcare professionals make informed decisions regarding their utilization in scientific publications and clinical practice.