It's Getting Way Too Easy to Create Fake Videos of People's Faces

Deepfakes continue to get better and better. Samsung researchers developed an algorithm that only needs one source image to create fake videos. By using as little as a single photo it is possible to create talking head videos by "[training] the program to identify what they call "landmark" features of the faces: eyes, mouth shapes, the length and shape of a nose bridge." This creates a 'new face' but one that is close enough such that in the 'scroll through the feed' world we consume content in, it gets harder to notice. And the tech is only going to get better.

The researchers write in the paper that they recognize the applications for realistic face-avatars in video conferencing, gaming, and special effects—but the uncanny valley often holds us back from fully embracing widespread use of face-avatars of real people. They hope that this work changes that, with its low source requirements and "perfect" realism.

With fake videos already making their rounds, like this one of Nancy Pelosi that spread through social channels this week that had been slowed down to make her speech seem slurred and garbled, this challenges faced with this technology are only going to magnify, and there doesn't seem to be a tech solution to this problem.


Want to receive more content like this in your inbox?