Large language models in ophthalmology: a scoping review on their utility for clinicians, researchers, patients, and...

Since its introduction in November 2022, the public interest in the utility of large language models (LLMs) has gained widespread adoption among individual consumers and among medical practitioners, with a consequent increase in publications describing their utility in healthcare. This review highlights original research articles on how LLM's can be utilized by various stakeholders in ophthalmology through clinical assistance, patient education, medical education, and research. ChatGPT consistently responds with better accuracy and quality than other LLMs across various studies employing different methodologies, with newer iterations offering more advantages. Studies have likewise identified limitations of LLMs, which include hallucination, inability to interpret image-based prompts, and limited performance across non-English languages. As newer iterations of available and more advanced models with image processing are currently being introduced, generative artificial intelligence should be continuously monitored for its implications in eye care.