LLM-assisted medical documentation: efficacy, errors, and ethical considerations in ophthalmology

Artificial intelligence (AI) has been recognised as a potentially transformative tool in modern medicine, with the ability to significantly enhance workflow efficiency [1]. Implementing AI to automate the writing of clinic notes is one area in which such benefit may be realised. Large language models (LLMs) are a subset of AI trained on vast amounts of textual data and have shown great promise in understanding and generating human-like text [2]. In ophthalmology, the integration of LLM-driven autocompletion functions introduces the potential for AI-generated management plans to be created. It is therefore important to consider their efficacy, reliability and potential to influence overall patient outcomes.