People in the AI world said video was a step to far, after seeing the insane leaps in image generation - compare this to the video AI generation of just a few MONTHS ago... insane - the "characters" are moving way more, the backgrounds are stable as well - heads and eyes turning and moving. These generators need to just take a little of the "gloves off" on certain things, and its done time.
That guy who was like, movies are over- you will just ask Alexa for a X type movie with Z people and Y type of plot and settings and boom - its not so crazy day by day.