AI chatbots outperformed dentists and doctors on radiation safety quiz

Artificial intelligence chatbots answered questions about radiation risks and safety more accurately than dentists, medical doctors, and members of the general public in a comparative study published in Scientific Reports. The findings, based on a 25-item multiple-choice questionnaire, showed that seven widely used AI systems achieved accuracy rates between 80 and 90.67 percent, while the 344 human participants scored between 32 and 60 percent. The research team, led by Hazal Karslıoğlu, Fatıma Sümeyra Çetin, and Halil Altun, administered the same questionnaire to dentists, physicians, and lay respondents, and then entered each question into seven selected AI-based chatbot systems: Claude, DeepSeek, Microsoft Copilot, Gemini, Perplexity AI, ChatGPT-5, and ChatGPT-4o.

The design allowed a direct comparison of response accuracy between human groups and machine systems on identical radiation safety items. Among the evaluated chatbots, Microsoft Copilot achieved the highest performance, while Perplexity AI exhibited the lowest. The spread between the best and worst AI systems was relatively narrow compared with the variation observed among human participants, whose accuracy spanned nearly thirty percentage points across the pooled groups.

However, the study found no significant difference between the two healthcare professional groups themselves, suggesting that radiation safety knowledge was comparable across the two professions under the conditions tested. The researchers also examined whether chatbot performance was consistent across different users and time periods. Most of the AI systems maintained stable accuracy when assessed in the morning, at midday, and in the evening.

ChatGPT-5 was the exception, showing a significant temporal difference in performance across the assessment windows, with a p-value below 0.001. This variability in one system highlights that consistency over time cannot be assumed for every AI tool, even when overall accuracy is high. Radiation exposure is encountered in daily life through both natural and artificial sources, as well as in medical diagnostic and therapeutic applications, and such exposure may lead to adverse health effects.

The authors concluded that AI-based systems achieved higher response accuracy than human participants under the evaluation conditions used in the study, and that the overall performance of the human participant groups indicates room for improvement in radiation safety education. The researchers were careful to frame the implications of their findings. While AI-based tools may serve as valuable educational resources for learning radiation safety concepts, the authors stated that their performance should not be interpreted as evidence of superior clinical judgment or decision-making ability.

The study, published as an open-access article on 26 October 2026, positions chatbots as a promising supplementary tool for radiation safety education rather than a replacement for professional expertise.

Subject of Research: Comparative assessment of radiation safety knowledge among AI chatbots, dentists, medical doctors, and the general public

Article Title: AI chatbots showed higher response accuracy than dentists, medical doctors, and the general public on radiation safety questions

Article References: karslıoğlu, H., Çetin, F. S., & Altun, H. (2026). AI chatbots showed higher response accuracy than dentists, medical doctors, and the general public on radiation safety questions. Scientific Reports. https://doi.org/10.1038/s41598-026-77832-8

Image Credits: AI Generated

DOI: 10.1038/s41598-026-77832-8

Keywords: AI chatbots, radiation safety, radiation risks, dentists, medical doctors, Microsoft Copilot, ChatGPT, Perplexity AI, questionnaire study, Scientific Reports, health education, response accuracy