A scoping review examined 10 studies to see how AI-generated videos are used to teach medical science. The researchers looked at the tools, prompts, and methods used to create these educational videos. They found that many of these projects lack a standard way to be measured or created.
One major finding was that half of the studies did not clearly document the specific instructions, or prompts, used to create the AI content. Additionally, when researchers graded the quality of these AI-generated videos using a standard scale, most received low scores of 2 out of 5. Only one study used three different tools to check for quality.
Because there is currently no standard way to make or judge these videos, it is hard to know if they are reliable for teaching. The evidence comes from a small number of studies and shows that more research is needed to create consistent rules for using AI in medical education.
