ChatGPT for tinnitus information and support: response accuracy and retest after three months

Abstract

BackgroundChatGPT – a conversational tool based on artificial intelligence – has recently been tested on a range of topics. However most of the testing has involved broad domains of knowledge. Here we test ChatGPT’s knowledge of tinnitus, an important but specialized aspect of audiology and otolaryngology. Testing involved evaluating ChatGPT’s answers to a defined set of 10 questions on tinnitus. Furthermore, given the technology is advancing quickly, we re-evaluated the responses to the same 10 questions 3 months later.Material and methodChatGPT (free version 3.5) was asked 10 questions on tinnitus at two points of time – August 2023 and November 2023. The accuracy of the responses was rated by 6 experts using a Likert scale ranging from 1 to 5. The number of words in each response was also counted, and responses were specifically examined for whether references were provided or whether consultation with a specialist was suggested.ResultsMost of ChatGPT’s responses were rated as satisfactory or better. However, we did detect a few instances where the responses were not accurate and might be considered somewhat misleading. The responses from ChatGPT were quite long (averaging over 400 words) and they occasionally tended to stray off-topic. No solid references to sources of information were ever supplied, and when references were specifically asked for the sources were artificial. For most responses consultation with a specialist was suggested. It is worth noting that after 3 months the responses generally improved.ConclusionsChatGPT provided surprisingly good responses, given that the questions were quite specific. Although no potentially harmful errors were identified, some mistakes could be seen as somewhat misleading. No solid references were ever supplied. ChatGPT shows great potential if further developed by experts in specific areas, but for now it is not yet ready for serious application.

Accuracy and Repeatability of ChatGPT Based on a Set of Multiple-Choice Questions on Objective Tests of Hearing

“…This study aimed to test repeatability using a set of standardized questions; however, we observed significant errors and variability in responses. This is particularly concerning when ChatGPT generates a narrative or answers an open-ended question without providing reliable sources or, in some cases, citing non-existent references [10]. The ability to track sources and verify responses is invaluable, particularly if responses can vary.…”

“…Preliminary studies in audiology suggest that while ChatGPT, alongside other chatbots like Google Bard (now Gemini) and Bing Chat (now Copilot), shows promise, it also exhibits errors and inaccuracies that underscore the need for careful oversight when used in specialized fields [9]. This is particularly evident in some audiology subtopics such as tinnitus, where the responses, although quite impressive, occasionally stray from the topic and, crucially, totally lack citations [10]. These latter two studies suggest that ChatGPT has the potential to provide information in more specialized medical fields like audiology and in specific topics like tinnitus, but still requires improvement before being reliable enough for serious applications.…”