Given the current barely useful state of automated captioning (on for example videos) - even when the speaker's voice is perfectly clear - I'm not optimistic th

Given the current barely useful state of automated captioning (on for example videos) - even when the speaker's voice is perfectly clear - I'm not optimistic that the accuracy will be up to the standards needed for a clinical setting. Plus it will basically encourage people to listen less (it will be in the report so why bother?), and just like that creepy box you have at home - it will always be listening, all the time. But hey, it's powered by AI so what's not to celebrate?