AI-Written Code Beats Human Experts in Medical Analysis

A study published in Cell Reports Medicine shows that large language models (LLMs) enable junior scientists to create highly accurate biomedical code ^[1]. The AI-generated code focused on predicting preterm birth risks through large-scale data analysis ^[1][2]. In several tests, the performance of these tools matched or even outperformed expert bioinformatician teams ^[1]. Experts suggest that AI tools could significantly lower the barriers to entry for complex medical research ^[1]. These models allow researchers with less technical training to achieve results previously reserved for specialists ^[1]. This shift may lead to faster advancements in maternal health and data-driven diagnostics ^[2].