Enrique Dans's article discusses the increasing trend of individuals using AI like ChatGPT to seek medical second opinions, sparked by anecdotes of AI outperforming doctors in diagnosis. 

This phenomenon is presented as an evolution of the "Doctor Google" syndrome, where people self-diagnose using online information. 

The author highlights a study indicating that ChatGPT's medical diagnostic accuracy is currently below 50%, suggesting successful instances might be outliers. A key distinction is made between a human doctor's diagnosis and an AI's, where the latter, based on its training, might consider a broader range of possibilities, even unlikely ones. 

The piece raises questions about the implications of patients potentially prioritizing AI-generated diagnoses over professional medical advice when such technology becomes more readily accessible.

This article is also available in English on my Medium page, «What happens when ChatGPT’s second opinion overrules your doctor’s?»