A team of Stanford University researchers developed benchmarks for measuring the accuracy and effectiveness of AI agents to assist physicians and published their findings in the New England Journal of ...