Table 2. Performance of ChatGPT in the second state examination (M2).
Subjects | Result (n, %, [95% CI]) | ||
AIEP | 8 | 75.0 | [36.3; 100] |
Ophthalmology | 7 | 85.7 | [50.8; 100] |
Surgery/orthopedics | 14 | 64.3 | [35.6; 93.0] |
Dermatology | 7 | 85.7 | [50.8; 100] |
Epidemiology | 15 | 46.7 | [18.1; 75.3] |
Gynecology | 20 | 80.0 | [60.8; 99.2] |
Otorhinolaryngology | 3 | 33.3 | [0; 100] |
Human genetics | 14 | 64.3 | [35.6; 93.0] |
Infectious diseases | 13 | 84.6 | [61.9; 100] |
Internal medicine | 34 | 64.7 | [47.8; 81.6] |
Neurology | 45 | 46.7 | [31.5; 61.8] |
Pediatrics | 23 | 65.2 | [44.2; 86.3] |
Pharmacology | 19 | 94.7 | [83.7; 100] |
Psychiatry | 9 | 66.7 | [28.2; 100] |
Radiology | 8 | 75.0 | [36.3; 100] |
Forensic medicine | 12 | 66.7 | [35.4; 98.0] |
Urology | 1 | 100.0 | / |
Total | 252 | 66.7 | [60.6; 72.2] |
AIEP, anesthesia, intensive care, emergency medicine, pain management/palliative care;
CI, confidence interval