Watermarking AI images to fight misinfo and deepfakes may be pretty pointless

Basically, it's 'not going to work'

Exclusive  In July, the White House announced that seven large tech players have committed to AI safety measures, including the deployment of watermarking to ensure that algorithmically-generated content can be distinguished from the work of actual people.…

ai images deepfakes deployment exclusive generated images people safety safety measures tech watermarking white house work