Accuracy and Same-Day Response Consistency of AI Chatbots on Complete Denture Multiple-Choice Questions

Aim To compare the accuracy and same-day response consistency of four free-tier AI chatbots on single-best-answer complete denture MCQs.Material and method Twenty-five English MCQs (A–E, single correct option) on complete denture prosthodontics were used. Four chatbots were tested via official web interfaces under default free-tier settings: ChatGPT, Claude, Gemini, and Grok. Each question was asked in a new chat to ensure zero prior context, using a standardized instruction requiring the output of only one option letter (“Answer: ”). Results Fleiss’ kappa showed significant within-day agreement for all chatbots (p