Prime Highlights-
- Eric Horvitz discusses AI’s growing impact on healthcare at Assembly Series talk.
- He highlights AI’s potential in diagnoses, treatment discovery, and protein biology research.
Key Facts-
- FDA cleared the first AI pathology testing tool for clinical use in 1995.
- Jen Nadler says AI governance depends on leaders and everyday users alike.
Background-
Microsoft Chief Scientific Officer Eric Horvitz spoke to a nearly full audience at Graham Chapel as part of the Assembly Series, sharing his views on how artificial intelligence is reshaping healthcare and medical research.
Horvitz traced his path at the crossroads of computer science and medicine to his time as an MD PhD student at Stanford University, where he learned that turning new technology into clinical practice takes patience. He noted that the Food and Drug Administration did not clear the first AI tool for pathology testing until 1995, years after early machine learning systems were first built.
Calling the recent leap in generative AI an inflection upon an inflection, Horvitz said the technology is opening doors to better clinical diagnoses, stronger healthcare decisions and faster treatment discovery. He cited breakthroughs in protein biology as proof of AI’s ability to tackle major challenges tied to cancer, neurodegeneration and autoimmune disease.
Even as AI expands its footprint in medicine, Horvitz stressed that human values and oversight cannot be set aside, adding that patient preferences and caregiver relationships need continued protection.
Jen Nadler, a portfolio manager at the Center for Health AI, said responsible governance will rely on everyone involved, from senior leaders to individuals using AI tools directly. PhD candidate Seyun Jeong and medical student Harshu Pande, who attended, voiced interest in applying AI to clinical outcomes and radiology. Horvitz ended his talk by urging the next generation to take an active role in shaping the technology.
