This is the first study to compare the readability and quality of AI-generated patient education materials compared to physician approved leaflets for common pediatric ophthalmic conditions. Patient information on four common eye conditions (chalazion, blepharitis, amblyopia, and strabismus) was generated using two large language models (ChatGPT-3.5 and DeepSeek R1) and compared with NHS-approved leaflets from Moorfields Eye Hospital. Readability was assessed using four validated tools: Flesch-Kincaid Grade Level (FKGL), Flesch Reading Ease (FRE), Gunning Fog Score, and SMOG index. Content quality was evaluated using the Ensuring Quality Information for Patients (EQIP) tool. Statistical analysis included ANOVA for group comparisons and Cronbach’s alpha to assess inter-rater reliability. NHS leaflets were significantly more readable than AI-generated materials (mean FKGL: 4.1 vs ChatGPT: 7.5, DeepSeek: 9.2). Only NHS resources met the recommended FRE score of 65 for the general population (mean FRE: NHS 74.8 vs ChatGPT 57.7, DeepSeek 48.1). EQIP scores showed significant differences in information quality with NHS leaflets scoring highest (mean = 69.61), outperforming ChatGPT (mean = 58.26) and DeepSeek (mean = 56.75). While AI tools offer accessible, updated health information, they fall short in readability and quality compared to clinician-authored resources. These limitations, along with a lack of references for information, pose risks for patients, particularly with limited health or digital literacy. Until these models reliably meet standards for clarity and accuracy, NHS-approved materials should remain the primary reference for patient education. With refinement and human oversight, AI has future potential for scalable, personalised, interactive health communication.

Dr. Haidar award-winning Specialised Academic Doctors training within the Imperial College Healthcare NHS Trust. uniquely hold two degrees from Imperial College London, an MBBS with Distinction and a First-Class Honours BSc in Endocrinology, reflecting exceptional clinical and scientific achievement. She have published in leading journals including The BMJ and Archives of Disease in Childhood and presented research at major national conferences such as RCPCH and RCPsych. She  also serve as editorial members for UK-based academic journals, contributing to the advancement of emerging medical research.