Netflix’s new reality series Deep Fake Love is based on the participants, five couples in serious relationships, showing how well they know each other. And the winning couple goes home with one million kroner. The couples are separated at the beginning of the series and instead have to live with singles. By being forced to watch videos of their partners cheating, they must then try to guess if what they see is real, or if it is only fake images, created by artificial intelligence.
International media have criticized the program for being Godless and Aftonbladet writes that “Love couples are subjected to AI torture”. Johan Öbrink is a digital strategist, and he too is very skeptical of the new series.
– I sat down and fast-forwarded through the season yesterday and part of my soul died yesterday, you could say, he says in Efter fem.
Can create revenge porn
Öbrink believes that there are certain problems with using AI in this way. Including the problem that you can now produce and spread a fake version of what is called revenge porn – that you save sex videos and leak them to take revenge on a former partner. Something they are trying to create legislation around.
– When a TV program does the same thing, it is clear that you start to push the boundaries of what is ethically correct, he says.
Not perfect
And despite the fact that the AI in the series is not perfect, we are starting to reach the level where it is difficult to distinguish between AI and reality, in the cases where the AI is well made, says Öbrink.
– Those actors who get their faces replaced are very similar to the people they are supposed to play. It makes it easier, but there are still “tells”. As humans, we can also pick up how to move the face and such, but that only applies if you know each other very well. We’re starting to reach the level where if it’s done well, it’s hard for the human eye to see what’s fake, he says.
According to Öbrink, however, in the case of Deep Fake Love, it is not the AI that creates the biggest drama.
– Of course it’s a new technology and a new level and see. And possibly the participants are manipulated by the production, it is possible. But Ain adds spice, it’s not the one who creates all the drama, he says.
Not just in reality
But it is not only in the reality world that we see the use of deep fake.
– A while ago, there were a number of EU parliamentarians who got a meeting booked with Navalny’s head of state and had a zoom meeting with him – who were not with him. Now they discovered it, but if you don’t have a really close relationship then it can be difficult to know if you’re talking to who you think you’re talking to.
Öbrink thinks we will see much more of this in the future.
– What is happening now is that what was previously very difficult to do is getting easier and easier, what you can do is getting better and better and the tools are becoming available to more and more people. So you don’t need to be very knowledgeable to create deep fakes yourself, he says.