Study: AI responses to healthcare queries are nearly 76% accurate
A study from Penn State reveals that AI responses to healthcare queries are approximately 76% accurate. While AI tools like ChatGPT can assist in answering health-related questions, the findings suggest that they are best utilized by trained physicians rather than patients. The research highlights the potential risks of relying on AI for medical advice, particularly in specialized fields.
- ▪AI-powered chatbots respond to health-related questions with nearly 76% accuracy.
- ▪The study involved a competition where participants used different large language models to generate health responses.
- ▪Specialties like obstetrics and gynecology performed better than internal medicine and neurology in terms of AI accuracy.
Opening excerpt (first ~120 words) tap to expand
ResearchCalling Doctor GPT: AI responses to healthcare queries are nearly 76% accurateArtificial intelligence shows promise for supporting physicians, but patient health questions are best left to human doctors, according to Penn State researchersLarge language models like ChatGPT respond to health queries with nearly 76% accuracy, raising concerns about their trustworthiness in real-world applications, according to Penn State researchers. Credit: fizkes/Getty Images. All Rights Reserved.ExpandMay 28, 2026By Francisco TutellaUNIVERSITY PARK, Pa.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at Psu.
Discussion
0 commentsMore from Psu
