Detecting AI‐Generated vs. Human‐Written Health Misinformation: The Impact of eHealth Literacy on Accuracy and Sharing

The widespread of AI‐generated health misinformation poses significant challenges to public health. This study aims to investigate the influence of eHealth literacy, the ability to seek, comprehend, and appraise health information from digital sources, on individuals' ability to detect AI‐generated (as opposed to human‐written) health misinformation and their subsequent sharing behaviors. We conducted an online experiment of 627 participants in which they were presented with 12 messages to detect both AI‐generated and human‐written misinformation. Results show that: 1) AI‐generated information is often perceived as more convincing, regardless of whether the information is true or false. 2) There is a significant negative correlation between eHealth literacy and detection accuracy, with younger and healthier participants exhibiting higher eHealth literacy but lower detection accuracy, highlighting the gap between self‐perception and actual eHealth skills. 3) Participants with lower accuracy in detecting misinformation are more likely to share health information and less likely to correct it when they are aware of its inaccuracy, thus spreading misinformation further. These findings highlight a disparity and challenge to use self‐perceived eHealth literacy as indicators of actual ability of handling health misinformation. Targeted interventions to enhance digital health literacy, particularly among younger and healthier populations, are urgently needed.