Table 2 The number and percentage of correct answers of chatbots.
Bard | Bing | GPT3.5 | GPT4 | |||
|---|---|---|---|---|---|---|
n (%) | n (%) | n (%) | n (%) | |||
Adult (1st and 2nd scenarios) | 1st response | 12 (60) | 7 (35) | 5 (25) | 17 (85) | |
2nd response | 10 (50) | 5 (25) | 5 (25) | 13 (65) | ||
Pediatric (3rd and 4th scenarios) | 1st response | 7 (30.43) | 10 (43.48) | 2 (8.7) | 5 (21.74) | |
2nd response | 12 (52.17) | 5 (21.74) | 7 (30.43) | 5 (21.74) | ||
Infant (5th and 6th scenarios) | 1st response | 2 (8.7) | 6 (26.09) | 5 (21.74) | 3 (13.04) | |
2nd response | 6 (26.09) | 2 (8.7) | 5 (21.74) | 2 (8.7) | ||
All scenarios | 1st response | 21 (31.82) | 23 (34.85) | 12 (18.18) | 25 (37.88) | |
2nd response | 28 (42.42) | 12 (18.18) | 17 (25.76) | 20 (30.3) | ||
