Accuracy and Reliability of Artificial Intelligence Chatbots as Public Information Sources in Implant Dentistry.

This article evaluates the accuracy, completeness, comprehensibility, and reliability of three widely available artificial intelligence (AI) chatbots—ChatGPT-3.5, Gemini, and Copilot—in responding to clinically relevant questions in implant dentistry. Twenty frequently asked questions, developed by experienced prosthodontists, were posed to each chatbot three times over 12-day intervals, and responses were independently scored using Likert scales and statistical analyses. Results showed no significant differences in overall accuracy among the chatbots, with Gemini demonstrating higher completeness but lower reliability compared to Copilot, which had the highest consistency. Despite acceptable levels of consistency, the study concludes that AI chatbots should not be solely relied upon for clinical decision-making in implant dentistry, emphasizing the necessity of prioritizing dental specialists' expertise.