The term Artificial Intelligence (AI) was first coined in the 1960s and has made significant progress up to the present day [1]. As of 2022, ChatGPT, developed by Open AI, has begun to find its place in our lives. The most current version, GPT-4, has been made available for use [2]. During this period, numerous AI applications have been developed. GPT-4 and Gemini are two of the best known of these AI models. [2,3]. AI applications developed in this process have begun to be used in various fields to make life easier. Medicine is one of these fields where AI is used, and numerous studies are being conducted related to GPT-4 and Gemini [4].

Accurate triage of patients arriving at the emergency department is a crucial component of emergency service delivery. High triage assessment of patients can lead to the misallocation of limited resources during periods when the emergency department is overcrowded. Furthermore, inadequate triage can pose a risk for adverse outcomes due to delays in diagnosis and treatment processes [5]. The Emergency Severity Index (ESI) is currently one of the most commonly used triage systems for effective patient triage in the emergency department. ESI categorizes patients into five categories based on the patient's clinical severity and resource utilization. Patients in ESI 1 (immediate) and ESI 2 (emergency) groups require urgent and critical care, while patients in ESI 3 (urgent), ESI 4 (semi-urgent), and ESI 5 (non-urgent) are clinically stable and classified according to the number of resources used [6]. Recently, there have been studies on the use of GPT-4 in triage [7,8]. In a study conducted by Naoki et al. investigating the ability of GPT-4 to triage typical clinical cases, it was concluded that GPT-4 exhibited a performance comparable to that of board-certified physicians in triaging patients [7]. Our literature review did not find any studies evaluating the performance of Gemini in the ESI triage system. Additionally, we anticipate that the lack of a study evaluating the ESI triage performances of GPT-4, Gemini, and emergency medicine specialists is a limiting factor for the usability of AI in the medical field.

The aim of this study is to evaluate the performance of GPT-4, Gemini, and emergency medicine specialists in ESI triage against each other; furthermore, it aims to contribute to the literature on the usability of these AI programs in emergency department triage.