Today’s article is about a new GAN (or General-Adversarial Network) based AI developed by Nvidia that can turn even simple, MS Paint looking doodles into works of photorealistic art, even if you’re an awful artist. This, obviously, can present some cause for concern for some people, namely in that art is assumed to be the One Big Thing™ that the AI can’t take away from us. Already, we have GANs capable of generating faces of people that don’t actually exist – give this site a whirl, and let us know how easily you could discern if those faces were AI generated or not. Either way, this is both an amazing step forward in neural network technology and a horrifying step forward into a post-photography era, where the concept of “photographic evidence” may no longer be a viable method of distinguishing truth from reality.
What do you think? I’d love to hear your thoughts on the subject below.