A comparative study on the knowledge levels of artificial intelligence programs in diagnosing ophthalmic pathologies and intraocular tumors evaluated their superiority and potential utility.

Abstract

This study aimed to test the knowledge levels of ChatGPT, Bing, and Bard artificial intelligence chatbots, which have been released by three different manufacturers, about ophthalmic pathologies and intraocular tumors, to test their usability and to investigate the presence of superiority to each other. Thirty-six questions were obtained from the American Academy and Ophthalmology 2022-2023 Basic and Clinical Science Course Ophthalmic Pathology and Intraocular Tumor study questions section. Each question was asked separately for the ChatGPT, Bing, and Bard artificial intelligence programs. Answers to the questions were categorized as correct or incorrect. The statistical relationship between the correct and incorrect response rates of the artificial intelligence programs was determined. From the artificial intelligence chatbots, ChatGPT gave the correct answer to 58.6% of the questions asked, Bing gave the correct answer to 63.9%, and Bard gave the correct answer to 69.4%. No statistical significance was found between the rates of correct answers to the questions in all 3 artificial intelligence programs (p = 0.705, Pearson Chi-square test). Artificial intelligence chatbots can be used to access information related to ophthalmic pathologies and intraocular tumors. However, in the evaluation of the data, it should be noted that not all questions can be answered correctly. Care should be taken when examining the answers.