Performance of GPT-5, DeepSeek, and Claude in dental MCQs for medically compromised patients

Puntos clave

• Accuracy in answering dental MCQs was tested, showing AI capabilities may assist dental education.
• GPT-5 achieved 85% accuracy, while DeepSeek and Claude followed closely with 80% and 75% respectively.
• The approach involved evaluating responses to multiple-choice questions based on clinical scenarios.
• Findings support the integration of AI tools in educational settings, underscoring their potential to enhance learning.