RESUMO
BACKGROUND: ChatGPT's (Chat Generative Pre-Trained Transformer) remarkable capacity to generate human-like output makes it an appealing learning tool for healthcare students worldwide. Nevertheless, the chatbot's responses may be subject to inaccuracies, putting forth an intense risk of misinformation. ChatGPT's capabilities should be examined in every corner of healthcare education, including dentistry and its specialties, to understand the potential of misinformation associated with the chatbot's use as a learning tool. Our investigation aims to explore ChatGPT's foundation of knowledge in the field of periodontology by evaluating the chatbot's performance on questions obtained from an in-service examination administered by the American Academy of Periodontology (AAP). METHODS: ChatGPT3.5 and ChatGPT4 were evaluated on 311 multiple-choice questions obtained from the 2023 in-service examination administered by the AAP. The dataset of in-service examination questions was accessed through Nova Southeastern University's Department of Periodontology. Our study excluded questions containing an image as ChatGPT does not accept image inputs. RESULTS: ChatGPT3.5 and ChatGPT4 answered 57.9% and 73.6% of in-service questions correctly on the 2023 Periodontics In-Service Written Examination, respectively. A two-tailed t test was incorporated to compare independent sample means, and sample proportions were compared using a two-tailed χ2 test. A p value below the threshold of 0.05 was deemed statistically significant. CONCLUSION: While ChatGPT4 showed a higher proficiency compared to ChatGPT3.5, both chatbot models leave considerable room for misinformation with their responses relating to periodontology. The findings of the study encourage residents to scrutinize the periodontic information generated by ChatGPT to account for the chatbot's current limitations.
RESUMO
BACKGROUND: Although Chat Generative Pre-trained Transformer (ChatGPT) (OpenAI) may be an appealing educational resource for students, the chatbot responses can be subject to misinformation. This study was designed to evaluate the performance of ChatGPT on a board-style multiple-choice dental knowledge assessment to gauge its capacity to output accurate dental content and in turn the risk of misinformation associated with use of the chatbot as an educational resource by dental students. METHODS: ChatGPT3.5 and ChatGPT4 were asked questions obtained from 3 different sources: INBDE Bootcamp, ITDOnline, and a list of board-style questions provided by the Joint Commission on National Dental Examinations. Image-based questions were excluded, as ChatGPT only takes text-based inputs. The mean performance across 3 trials was reported for each model. RESULTS: ChatGPT3.5 and ChatGPT4 answered 61.3% and 76.9% of the questions correctly on average, respectively. A 2-tailed t test was used to compare 2 independent sample means, and a 2-tailed χ2 test was used to compare 2 sample proportions. A P value less than .05 was considered to be statistically significant. CONCLUSION: ChatGPT3.5 did not perform sufficiently well on the board-style knowledge assessment. ChatGPT4, however, displayed a competent ability to output accurate dental content. Future research should evaluate the proficiency of emerging models of ChatGPT in dentistry to assess its evolving role in dental education. PRACTICAL IMPLICATIONS: Although ChatGPT showed an impressive ability to output accurate dental content, our findings should encourage dental students to incorporate ChatGPT to supplement their existing learning program instead of using it as their primary learning resource.
