Deepfake app ZAO realistic, terrifying
‘‘China’’, ‘‘Deepfake’’ and ‘‘Privacy’’ are three of the most controversial themes of 2019. And now all three themes can be found neatly wrapped up in a single app.
ZAO is a new app from China that lets users create deepfake videos in a matter of seconds.
The app was released last Friday and its immediate popularity has seen it dominate China’s domestic download charts.
The app instructs users to upload a series of selfies where they repeat basic facial movements such as blinking, moving their mouths and acting out several other emotions.
Once the app has these videos it can realistically superimpose a user’s face onto scenes from
television shows and movies.
The results are as realistic as they are scary.
Equally terrifying was the app’s privacy policy that initially stated that any uploaded content granted Momo, the company that developed the app, ‘‘free, irrevocable, permanent, transferable, and relicenseable’’ rights to all usergenerated content’’.
Since then, the ZAO app has released the following statement: ‘‘We understand the concerns about privacy. We’ve received the feedback, and will fix the issues that we didn’t take into consideration, which will take some time.’’
Deepfake videos aren’t new. In June an intimidating deepfake video of Mark Zuckerberg emerged where his dubbed face relayed the following chilling monologue: ‘‘Imagine this for a second: One man, with total control of billions of people’s stolen data, all their secrets, their lives, their futures. I owe it all to Spectre. Spectre showed me that whoever controls the data, controls the future.’’
Apps that require users to voluntarily upload images of their faces are equally unoriginal.
Earlier in the year, the Russianmade FaceApp topped global download, despite requiring its users to handover HD close-up images of their faces.
The use of controversial facerecognition software is on the rise globally. And, no matter how cool the app might be, users might want to err on the side of caution before handing over personal data – which, in this case, would be images of their face.