Post by Mass General Brigham

How well do today's AI models perform in real-world clinical settings? Researchers at Mass General Brigham developed BRIDGE, a multilingual benchmark, to see how well large language models (LLMs) understand real-world clinical language across nine languages. The study found a significant gap between AI performance on standardized medical exams and real-world clinical tasks. By providing a more realistic way to evaluate medical AI, BRIDGE has the potential to help advance more accurate and equitable AI tools for non-English-speaking patients. Learn more: http://spklr.io/6040EOdx6