Originally published in Newsday by Will Coggin on April 1, 2019:
There’s a cutting-edge trend in video development with promising — and alarming — implications. This double-edge sword is known as “deepfake,” and it uses artificial intelligence to create convincing videos of people doing and saying things they haven’t actually done. Deepfake gained notoriety after users on the popular internet forum Reddit created fake pornographic videos of celebrities, though it has also seen more benign uses.
Deepfake technology comes with potential for propaganda that could easily be capitalized on by bad actors. Imagine a video of a presidential candidate saying or doing something that would evaporate their support — or manipulated footage of a public figure doing something illegal. It’s a matter of when, not if, deepfake technology will be used for propaganda purposes. The only antidote is Americans must adopt a deeper level of scrutiny and resist urges to rush to judgment.
Video has long been used as a tool to advance agendas — one need only think of Nazi film director Leni Riefenstahl. Even ISIS has a propaganda arm and will do multiple takes and use scripts, staging battles and other events in order to maximize the effectiveness of the group’s propaganda.
Staged propaganda can also be used in the service of lobbying campaigns. An explosive report published this month by fashion magazine Women’s Wear Daily revealed that, for over a decade, animal rights activists at PETA and other groups have been using a staged video to drum up support for anti-fur campaigns.