Once upon a time, faking faces in photographs required a high level of skill and talent, as well as a large amount of time. But more than 15 years ago, mankind had such a tool as Photoshop and making fake from the source has become much easier. It seemed that it was impossible to do such a fraud on video, but a team of engineers from the USA and Germany developed the Face2Face program for changing faces in real time to change the speaker’s facial expressions and speech to the desired one. As the developers assure, an actor (or who something else), the facial expression of which will be superimposed on the face of the “victim”, a standard RGB camera, a computer with a processor no worse than Intel Core i7 and an Nvidia GTX980 graphics card. Using the image depth sensor, the program creates a mask for each participant in which there are bindings to certain facial points on the faces. After that, the system on the fly creates a realistic video version of the face and puts it on top of the speaker’s real face, while the position of his head does not play a role at the moment. The capabilities of Face2Face can be assessed in the attached video, in which the developers took advantage of the performances of world famous figures (George W. Bush, Vladimir Putin, Donald Trump and Barack Obama). This technology is not available outside the research laboratory, but only for now. As you can see from the video, in some places the overlay betrays itself with artificial and “strained” facial expressions. The program is being finalized, but the creators hope for its commercialization in the near future. Face2Face can be used for creating films and realistic computer games, for dubbing videos into different languages, and so on. I would like to believe that things will not come to political fakes and profanities.