sabreW4K3@lazysoci.al to Technology@beehaw.org · 4 months agoAstronomers discover technique to spot AI fakes using galaxy-measurement toolsarstechnica.comexternal-linkmessage-square25fedilinkarrow-up1121
arrow-up1121external-linkAstronomers discover technique to spot AI fakes using galaxy-measurement toolsarstechnica.comsabreW4K3@lazysoci.al to Technology@beehaw.org · 4 months agomessage-square25fedilink
minus-squareAndrasKrigare@beehaw.orglinkfedilinkarrow-up7·4 months agoLooking completely realistic and being able to discern between real and fake are competing goals. If you can discern the difference, then it does not look completely realistic. I think what they’re alluding to is generative adversarial networks https://en.m.wikipedia.org/wiki/Generative_adversarial_network where creating a better discriminator that can detect a good image from bad is how you get a better image.
Looking completely realistic and being able to discern between real and fake are competing goals. If you can discern the difference, then it does not look completely realistic.
I think what they’re alluding to is generative adversarial networks https://en.m.wikipedia.org/wiki/Generative_adversarial_network where creating a better discriminator that can detect a good image from bad is how you get a better image.