Researchers used an OpenAI model to reanalyze 376 unsolved genetic disease cases in children, leading to 18 new diagnoses. The model helped surface leads by connecting clues from genomic sequencing, clinical records, and scientific literature. This approach shows promise for tackling the challenging task of diagnosing rare genetic diseases, where roughly half of cases remain unsolved despite extensive testing. By leveraging AI to reexamine existing data, experts can uncover new insights and make diagnoses that were previously impossible, which is significant because it demonstrates how AI can help accelerate the diagnosis of rare diseases, ultimately improving patient outcomes and Why it matters: this study highlights the potential for AI to revolutionize the field of rare disease diagnosis by making it more efficient and effective.