Adams explains digital trickery in Mozilla’s ‘Misinfo Monday’

NLP in the News


Peter Adams explains digital trickery, including deepfakes and cheapfakes, in the Mozilla Foundation’s Aug. 31 installment of Misinfo Monday: Deepfakes and Other Trickery in Imagery.

“Deepfakes are algorithmically-manipulated digital assets,” Adams says. “Deepfakes can be video or audio or even, nowadays, just an image.”

While deepfakes are computer-generated, they are not the same as the  computer generated imagery (CGI) found in many films. “CGI that movies use is imagery that is entirely fabricated in post-production by digital artists,” he says. “A deepfake, on the other hand, uses an algorithm that has learned how someone’s face looks and moves, and maps that onto authentic footage.”

But far more common on the internet are easier-to-create cheapfakes. “A cheapfake is a video or image that simply gets taken out of context,” Adams explains. “For example, taking an old photo of a crowd and saying it was an anti-Covid 19 protest. Or the crudely doctored video of slowed-down Nancy Pelosi. Cheapfakes are incredibly easy to do, since they generally only require you to copy/paste.”

 

 

More Updates

Washington Post’s Valerie Strauss showcases The Sift

Washington Post education reporter Valerie Strauss features content from The Sift®,  NLP’s free weekly newsletter for educators, in her blog throughout the school year. Using ChatGPT for disinformation and other news literacy lessons (Feb. 2, 2023 Lisa Marie Presley’s death, AI problems and more news literacy lessons (Jan. 25, 2023) News literacy lessons: ‘Shark Week,’…

NLP in the News