Artificial intelligence (AI) has the potential to transform the future of health care, but according to Dr. Ziad Obermeyer, it can also harm as much as it can help. That’s because bias and errors are built into the very algorithms we use to predict health care needs, “reproduce[ing] all of those ugly things that we don’t like about our health care system.” In this episode, we speak with Dr. Obermeyer about the hidden signals in health care data, his research revealing racial bias in health care AI and how we can fix algorithms so that machine learning can be a force for good.

Guest: Ziad Obermeyer, MD, Blue Cross of California Distinguished Associate Professor of Health Policy and Management, UC Berkeley School of Public Health

Tags: AI, artificial intelligence, Business Group on Health, Dr. Ziad Obermeyer, health care bias, health care data, health equity, machine learning
 
Posted in: Artificial Intelligence, Audio Podcast, Business Group on Health, Healthcare