\n<\/p>\n
Google will soon start requiring<\/a> political advertisers to \u201cprominently disclose\u201d when they made their ads with AI, as reported earlier by Bloomberg<\/em><\/a>. Starting in November, Google says advertisers must include a disclosure when an election ad features \u201csynthetic content\u201d that depicts \u201crealistic-looking people or events.\u201d<\/p>\n<\/div>\n That includes political ads that use AI to make someone look as if they\u2019re saying or doing something they never did, as well as changing the footage of an actual event (or fabricating a realistic-looking one) to create a scene that never happened.<\/p>\n<\/div>\n Google says these types of ads must contain a disclaimer in a \u201cclear and conspicuous\u201d place, noting that it will apply to images, videos, and audio content. The labels will need to state things like, \u201cThis audio was computer generated,\u201d or \u201cThis image does not depict real events.\u201d Any \u201cinconsequential\u201d tweaks, such as brightening an image, background edits, or removing red eye with AI, won\u2019t require a label.<\/p>\n<\/div>\n \u201cGiven the growing prevalence of tools that produce synthetic content, we\u2019re expanding our policies a step further to require advertisers to disclose when their election ads include material that\u2019s been digitally altered or generated,\u201d a Google spokesperson Allie Bodack says in a statement to The Verge.<\/em><\/p>\n<\/div>\n