🩺 Medical AI may rely on pattern matching, not reasoning
When common answer patterns were disrupted, all tested large language models saw accuracy drops, raising concerns about readiness for clinical decision-making.
🔗 doi.org/10.1001/jama...
#SciComm 🧪 #MedicalAI #HealthTech #GenAI
doi.org
Fidelity of Medical Reasoning in Large Language Models
This cross-sectional study evaluates whether the performance of large language models on medical benchmarks reflects logical reasoning or pattern recognition.
5 17 44
