A pilot evaluation of a new tool to evaluate generative AI outputs in healthcare has found that clinical assessors were dissatisfied with referencing, accuracy and the ability to recognise uncertainty.
A pilot evaluation of a new tool to evaluate generative AI outputs in healthcare has found that clinical assessors were dissatisfied with referencing, accuracy and the ability to recognise uncertainty.
This content is for paid subscribers only.
Start your free trial nowor purchase a monthly or annual subscription to read.