Why prompting matters: achieving clinically accurate and consistent responses with Chat <scp>GPT</scp>

Why prompting matters: achieving clinically accurate and consistent responses with Chat GPTArtificial intelligence (AI) has become a powerful tool in healthcare, offering the potential to analyse vast amounts of online information and provide personalised responses to user inquiries including patient education on their health.However, patient trust in AI-generated medical information remains a significant concern.A survey conducted by the University of Michigan revealed that 74% of adults aged >50 years did not trust AI-generated health information, while 68% were having trouble finding health information online [1].This concern underscores the necessity to evaluate AI-generated responses to ensure they do not perpetuate misinformation.