Skip to main content
. 2024 Jun 26;24:694. doi: 10.1186/s12909-024-05630-9

Table 2.

Classification of AI chatbots’ correct answers

N not applicable Logical reasoning
N(%)
Internal information
N(%)
External information
N(%)
N tot valid answers
ChatGPT-4 12 622 81.52 141 18.47 0 0 763
Microsoft Copilot 0 405 54.95 137 18.58 195 26.45 737
Google Gemini 12 567 98.78 7 1.21 0 0 574

N number, % percentage