Asking medical questions of AI with language spoken “in the wild”—meaning with LLM prompts from everyday consumers—brings back answers with decidedly mediocre accuracy, a new study shows.
Black and Hispanic individuals—known to be at greater risk of developing the degenerative disease—are significantly less likely than others to show Alzheimer’s pathology on brain scans, according to new research.
Asking medical questions of AI with language spoken “in the wild”—meaning with LLM prompts from everyday consumers—brings back answers with decidedly mediocre accuracy, a new study shows.