In a new study, artificial intelligence (AI) outperformed glaucoma specialists and matched retina specialists in diagnostic and treatment accuracy.

The study, published in JAMA Ophthalmology, compared a large language model (LLM) chatbot’s responses with those of eye specialists. In 2023, ChatGPT-4 was used as a medical assistant in New York eye clinics to answer glaucoma and retina questions, as well as in de-identified glaucoma and retinal cases.

The combined question-case mean rank for accuracy was 506.2 for the LLM chatbot and 403.4 for glaucoma specialists. Researchers concluded that their findings support the possibility that AI tools could play a pivotal role as both diagnostic and therapeutic adjuncts.