Table 1 –

Overall and age-stratified average reviewer ratings of GPT-4 and Gemini across five evaluation criteria

Large Language Model	Accuracy, mean (95% CI)	Completeness, mean (95% CI)	Age-Appropriateness, mean (95% CI)	Possibility of Demographic Bias, mean (95% CI)	Overall Quality, mean (95% CI)
GPT-4	4.37 (4.27, 4.47)	4.25 (4.16, 4.34)	3.95 (3.81, 4.09)	1.61 (1.49, 1.73)	3.88 (3.75, 4.01)
Gemini	4.55 (4.45, 4.65)	4.39 (4.28, 4.50)	3.26 (3.09, 3.43)	1.16 (1.11, 1.21)	3.43 (3.26, 3.60)
P-value	0.08	0.15	<0.001	<0.001	0.004

CI = confidence interval