A Harvard study found that AI systems outperformed human doctors in emergency medicine triage, achieving higher diagnostic accuracy during critical moments. Independent experts described the results as a significant advancement in clinical practice.

AI-powered carb counting tools can provide inconsistent estimates, yielding different results even when given the same input. This variability poses significant risks for individuals managing diabetes, as inaccurate carb counts can lead to hypoglycemic emergencies.

GPT-Rosalind is a new AI model designed to accelerate scientific research and drug discovery in life sciences. It is optimized for workflows in biology, chemistry, protein engineering, and genomics, enhancing tool use and understanding in these fields.