I wonder how far we are from the phones running some sort of Stable Diffusion AI with the photo taken as an input to create various fixed and touched up scenes.
One touch removal of people and background intrusions (even goes as far as suggesting items for removal).
So, not far I imagine!