Background: Complementary feeding is essential for infant growth, development, and health, but parents often rely on online sources for guidance, which may lack accuracy. ChatGPT, a widely used AI language model, has potential as an accessible source of health information. This study evaluates the accuracy and reproducibility of ChatGPT-4’s responses to common questions about complementary feeding. Methods: A prospective study was conducted using ChatGPT-4 to answer 20 frequently asked questions about complementary feeding. Responses were evaluated independently by two reviewers, with discrepancies resolved by a third reviewer. Answers were categorized as ”comprehensive,” ”correct but insufficient,” ”some correct or incorrect,” or ”completely incorrect.” Statistical analysis was conducted to assess inter-rater reliability and response reproducibility. Results: ChatGPT-4 provided ”comprehensive” answers for 75% of questions and ”correct but insufficient” answers for 20%, with no responses deemed partially or completely incorrect. Inter-rater reliability was high (Cohen’s kappa: 0.859). The AI model demonstrated consistent and reproducible performance across evaluations. Conclusion: ChatGPT-4 provides accurate and reproducible information on complementary feeding, aligning with professional standards. While promising as a supplementary information tool, healthcare professionals’ oversight remains critical. Further research is needed to optimize AI’s role in patient education and clinical practice.