The Future of AI: Enhancing Patient Safety

Generative artificial intelligence (Gen AI), particularly GPT-4, is making strides in healthcare, especially in patient safety, Earth.com revealed today.

A recent study from the BU Chobanian & Avedisian School of Medicine tested GPT-4’s proficiency using a 50-question exam from the Certified Professional in Patient Safety (CPPS). Remarkably, the AI answered 88% of the questions correctly without any prior medical training.

Dr. Nicholas Cordella, a co-author of the study, highlighted that this marks the first robust evaluation of AI in patient safety. The findings suggest AI could help clinicians identify and prevent medical errors, which are linked to approximately 400,000 deaths annually.

While the potential benefits are significant, experts caution against biases and inaccuracies inherent in AI models. Ongoing research is vital to refine these technologies for clinical use.