The Pentagon’s Defense Advanced Research Projects Agency is working with universities and research collaborations around the country to combat deep fake technology. Deep fakes pose a unique and near-existential threat to countries like the U.S. where free media and fair elections are an important cornerstone of civic life. Deep fake media driven by A.I., however, threatens to undermine trust in media and other public institutions, causing conspiracy theories to proliferate, as well as provide disinformation campaigns with a tool that could be easily used to decide an election. While Hollywood and other sectors have advanced human-driven digital video editing capabilities, deep fakes rely on artificial intelligence that could nearly perfect the art and allow malicious actors to release fake videos or audio recordings of politicians or others in fabricated situations. If these deep fakes cannot be spotted, then trust in visual or audio evidence could be completely undermined, removing a one of the foundations of the way in which reality and truth has been identified for the past 100 years, from the holocaust to moon landings, from Nixon’s “smoking” gun” audio tape to the Trump bus recordings. In this CNN interactive article, one is able to view several leading deep fake examples, as well as attempt to pick out deep fake samples from their original source videos.
Source: Deepfake videos: Inside the Pentagon’s race against disinformation
