This study evaluates the ability of artificial intelligence (AI)-powered tools, including ChatGPT-3.5, ChatGPT-4o (o1 Preview), and Google Gemini, to generate and improve the readability of existing educational materials on childhood myopia.

It demonstrates that these models, particularly ChatGPT-4o (o1 Preview), can produce good-quality, understandable, accurate, user-friendly content that meets readability standards, and improves the readability of existing online educational materials on childhood myopia at or below the sixth-grade level.

The study also highlights the importance of prompt customization in enhancing content clarity and provides practical recommendations for leveraging AI tools to improve health education.

The study identifies areas for improvement, such as enhancing actionability and incorporating multimedia elements.