Seems like this team from Stanford, the Max Planck Institute for Informatics and the University of Erlangen-Nuremberg, took the phrase ‘ Putting words into somebody’s mouth‘ too seriously and actually managed to do it.

Because the team produced a video demonstrating how its software, called Face2Face, in combination with a common webcam, can make any person on video appear to say anything a source actor wants them to say, reported Mashable. So, this facial recognition software tracks a subject and mirrors expressions on a conventional YouTube video, in real time. 

Simply put, the words will be yours but someone else’s mouth will be the medium, or should we call it another form of veiled puppetry? 

For the demo video, which is making the rounds on social media, the team used YouTube videos of U.S. President George W. Bush, Russian President Vladimir Putin and candidate  for Republication nomination for US President, Donald Trump. 

“Our goal is to animate the facial expressions of the target video by a source actor and re-render the manipulated output video in a photo-realistic fashion,” the researchers told Wired.

Watch this video which will give you a better understanding of the whole process. 

 Video Source: Matthias Niessner/YouTube

Why are we calling it scary?

Watching the video was was fun, right? But think how the misuse of this technology can have terrible repercussions. You could have videos doctored and incriminate anyone. As of now, the researchers have not made this particular application public but are reportedly thinking about commercializing it. If they do so, gone will be the days where you could at least trust a video (even partially). How will we believe anything anymore? 

Besides that, the chaos and rancour caused in India after doctored videos were aired by prominent news channels, accurately demonstrates the kind of damage such technology is capable of.