Evaluation of the Accuracy of Composite Shade Selection: Artificial Intelligence Chatbots Versus a Dentist

Abstract

Background: This study aimed to evaluate and compare the accuracy of composite shade selection between 2 artificial intelligence (AI) chatbots, ChatGPT-4.0 and Gemini Advanced, and a dental specialist.

Methods: A total of 21 crowns with different shades were fabricated to simulate various clinical scenarios. Three shade tabs were placed on each crown, and standardized photographs were taken using a smartphone. The shade selections of ChatGPT-4.0 and Gemini Advanced, based on standardized photographs, were compared to the visual shade evaluations conducted by a dental specialist. The groups were compared in terms of color difference (ΔE00) values and shade selection accuracy. Data were analyzed using 1-way ANOVA, Tukey’s honestly significant difference post-hoc test, and Pearson’s chi-square test.

Results: No statistically significant differences were observed among the groups in terms of either color difference values or accuracy of shade selection (P > .05). The shade selection accuracy was 57.1% for the dental specialist, 47.9% for Gemini Advanced, and 42.9% for ChatGPT-4.0.

Conclusion: Although the dental specialist demonstrated a higher success rate in shade selection, the differences were not statistically significant.

Cite this article as: Özdemir SB, Özdemir B. Evaluation of the accuracy of composite shade selection: artificial intelligence chatbots versus a dentist. Essent Dent. 2026, 5, 0084, doi:10.5152/EssentDent.2026.25084.