In real-world test, an AI model did better than ER doctors at diagnosing patients
A study published in Science found that an AI model developed by OpenAI outperformed doctors in diagnosing patients and making decisions about their care. The AI model was tested on real-world cases and excelled at providing accurate diagnoses using electronic health records.
Researchers at Harvard Medical School and Beth Israel Deaconess Medical Center tested an AI model developed by OpenAI on real-world cases. The AI model outperformed two experienced physicians in diagnosing patients and making decisions about their care. The study used electronic health records and limited information available to physicians at the time. The AI model excelled at providing accurate diagnoses, even in tricky cases. The researchers emphasize that the study relied on text alone, whereas clinicians use multiple inputs like images and nonverbal cues. The study's findings suggest that AI technology has advanced significantly and may be ready for clinical use.
This content was automatically generated and/or translated by AI. It may contain inaccuracies. Please refer to the original sources for verification.