So we know publicly that faking someone’s voice digitally is doable, if imperfect (there was a Jordan Peele video (TED talk?) where he pretended to be Obama to demonstrate this. And deepfakes for video.
Privately, is the software (technology?) good enough that it would be undetectable? Like, alphabet agency level software? I just assume so, but does anyone have any kind of informed opinion?
True, but I was more just curious about the technical capability.
I agree completely. I was just wondering if anyone had any sauce. I probably should have started with that.
With deepfakes, there’s an article from the MIT Technology Review from 2000 that demonstrates live and significant manipulation of videofootage. Nascent as shit, but chilling given how old that tech is.
Lying with pixels.
https://www.technologyreview.com/s/400734/lying-with-pixels/