This is why that ML/AI realtime deepfake software is so scary. Imagine if the person speaking is just mapping their facial expressions onto the face of the pers

This is why that ML/AI realtime deepfake software is so scary. Imagine if the person speaking is just mapping their facial expressions onto the face of the person you're actually getting. >This is why remote exams have all of those strict requirements like "show us your room" and "don't leave sight of the webcam" I'd imagine this could be defeated with a prerecorded video of the "interviewee" showing their room.