Google’s Med-PaLM 2, an artificial intelligence tool designed to answer questions about medical information, has been in testing at the Mayo Clinic, among others, since April. Med-PaLM 2 is a variant of PaLM 2, which was announced at Google I/O in May of this year. PaLM 2 is the language model that powers Google’s Bard.
See also: Google Play: Popular file manager apps were spyware
The Wall Street Journal (WSJ) reported that an internal email it saw indicated that Google believes updated model could be particularly beneficial in countries with “limited access to doctors.” Med-PaLM 2 was trained using a carefully selected set of medical professional demonstrations, which Google believes will make it more adept at healthcare conversations than generalized chatbots like Bard, Bing , and ChatGPT.
See also: How to enable the Google Maps Timeline feature for your account
The paper also cites research published by Google in May (PDF) that shows that Med-PaLM 2 still suffers from some of the accuracy issues we’ve come to expect from large language models. In the study, doctors found more inaccuracies and irrelevant information in the answers provided by Google’s Med-PaLM and Med-PalM 2 than those provided by other doctors.
However, on almost every other measure, such as demonstrating evidence of reasoning, supporting answers with consensus, or not showing signs of misunderstanding, Med-PaLM 2 performed roughly the same as real doctors.
The Wall Street Journal reports that customers testing Med-PaLM 2 will have control over their data, which will be encrypted, and Google will not have access to it.
See also: How to use picture-in-picture mode with Google Meet
According to Greg Corrado, a senior research manager at Google, the Wall Street Journal (WSJ) reports that Med-PaLM 2 is still in its early stages.
Source of information: theverge.com
