Comparative Evaluation of Reliability, Quality, Readability, and Originality of AI Chatbot Responses to Patient Questions in Implant Dentistry

Introduction/Objective This study aimed to evaluate and compare the reliability, informational quality, readability, and originality of responses generated by three widely used artificial intelligence chatbot platforms (Google Gemini Flash, ChatGPT Plus (OpenAI), and Claude Sonnet (Anthropic)) to patient-oriented questions in implant dentistry. Methods Twenty frequently asked questions in implant dentistry were developed and categorized as theoretical (n = 10) or clinical (n = 10). Each question was submitted to the three chatbots on the same day, and only the first response was analyzed. Responses were evaluated for reliability, informational quality, readability, and originality using validated assessment tools. Statistical comparisons were performed using appropriate parametric and nonparametric tests, with significance set at 0.05. Results Google Gemini Flash demonstrated significantly higher reliability scores than ChatGPT Plus and Claude Sonnet ( p < 0.001). Informational quality also differed significantly among models ( p = 0.045), with Gemini providing higher-quality responses than Claude. Readability analyses showed that chatbot responses generally required a high school or college-level reading ability. Similarity index values indicated high originality, with no significant differences among platforms ( p = 0.423). Differences were more pronounced for theoretical questions, whereas clinical questions showed limited variation. Discussion Although AI chatbots can provide useful information for implant dentistry patients, variability in reliability and readability highlights the need for cautious use without professional supervision. Conclusion AI-powered chatbots can support patient education in implant dentistry; however, response quality and accessibility vary across models. Expert oversight remains essential, and future research should focus on improving readability, transparency, and standardized evaluation of AI-generated dental information.