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

Table 3.

Classification of AI chatbots’ errors

N not applicable Logical error
N(%)
Information error
N(%)
Statistical error
N(%)
N total errors
ChatGPT-4 12 40 88.98 4 4,50 1 22.22 88.88
Microsoft Copilot 0 66 79.01 9 11.39 8 70.23 79.01
Google Gemini 12 233 99.57 1 1.00 0 0 99.57

N number, % percentage