Aims: In addition to the popular search engines on the Internet, ChatGPT may provide accurate and reliable health information. The aim of this study was to examine whether ChatGPT's responses to frequently asked questions concerning cerebral palsy (CP) by families were reliable and useful.
Methods: Google trends were used to find the most frequently searched keywords for CP. Five independent physiatrists assessed ChatGPT responses to 10 questions. Seven-point Likert-type scales were used to rate information reliability and usefulness based on whether the answer can be validated and is understandable.
Results: The median ratings for reliability of information for each question varied from 2 (very unsafe) to 5 (relatively very reliable). The median rating was 4 (reliable) for four questions. The median ratings for usefulness of information varied from 2 (very little useful) to 5 (moderately useful). The median rating was 4 (partly useful) for seven questions.
Conclusion: Although ChatGPT appears promising as an additional tool for informing family members of individuals with CP about medical information, it should be emphasized that both consumers and health care providers should be aware of the limitations of artificial intelligence-generated information.
Keywords: Artificial intelligence; ChatGPT; cerebral palsy; health information; large language model.