Google’s Med-PaLM 2, an artificial intelligence (AI) tool designed to provide answers to medical queries, has been undergoing testing at prestigious medical institutions, including the Mayo Clinic research hospital, since April. This variant of PaLM 2, initially introduced at Google I/O in May, serves as the language model behind Google’s Bard.
Improved Access to Medical Information:
According to a report by The Wall Street Journal, an internal email from Google highlighted that the updated Med-PaLM 2 model could prove particularly valuable in regions with limited access to doctors. Med-PaLM 2’s training involved a curated collection of medical expert demonstrations, which Google anticipates will enhance its capabilities for healthcare-related conversations compared to more generalized chatbots like Bard, Bing, and ChatGPT.
Comparable Performance in Key Metrics:
However, in almost every other metric evaluated, such as demonstrating reasoning abilities, providing consensus-supported answers, and exhibiting correct comprehension, Med-PaLM 2 performed on par with or better than human doctors. Despite the accuracy challenges, the chatbot shows promise in delivering reliable medical information.
Google’s Med-PaLM 2 AI chatbot is currently undergoing testing in renowned hospitals, including the Mayo Clinic. While the model continues to grapple with accuracy issues, it showcases notable performance in various metrics and holds potential in countries where access to healthcare professionals is limited. As development progresses, Med-PaLM 2 could become a valuable tool in providing accessible medical information and supporting healthcare conversations.