
NewsMore in News→
Harvard-Led Study Finds One OpenAI Model Outperformed Physicians in Early ER Diagnosis
Key Takeaways
- A Harvard-led study compared OpenAI models with attending physicians on 76 real ER cases.
- The o1 model produced the exact or a very close diagnosis in 67% of triage cases.
- Researchers said the models received the same unprocessed electronic-record information available to clinicians.
- The findings support AI as a potentially powerful diagnostic support tool rather than a replacement for doctors.
DE
DT Editorial Team··via techcrunch.com