ChatGPT-4 in Ophthalmology: the Third Wave of Artificial Intelligence

Abstract

The aim of this study is to explore the impact of ChatGPT-4 and other large language models in ophthalmology, highlighting their potential to revolutionize detection, diagnosis, and education in this field. Through a recent literature review and analysis of unimodal and multimodal studies, it is evidenced that ChatGPT-4 could improve diagnostic support capabilities in ophthalmology, surpassing previous tools and offering an accessible approach to the integration of artificial intelligence into clinical practice. Pedagogical studies also show its efficacy in medical education, albeit with limitations compared to human expertise. Ethical challenges and the need for adequate regulation are critical to its effective implementation. This study underscores the importance of “cue engineering” and customization of large language models for specific uses in ophthalmology, highlighting both the transformative potential and challenges associated with the adoption of conversational artificial intelligence in this field.