Deep Fake Love: How Netflix makes profit out of deepfake defamation

Is your significant other cheating on you, or have you just seen an AI-generated video? That’s how a new Netflix reality show is gambling with human emotions. While the entertainment industry is catching up with the tech buzz, isn’t it bringing the worst out of deepfake technology by using fake sexual content to defame?

Critics of Reality TV shows call it a spectacle of humiliation. While the drama, tears, and conflicts grab the audience's attention more than intoxicatingly than ever, the new Spanish reality TV show on Netflix, “Deep Fake Love,” brings ‘a spectacle of humiliation’ to a whole new level.

Despite the name, reality TV has always been mired in controversy and suspected of being fake. The footage on these shows is usually heavily edited, and the “real” participants may be closely instructed on what to do and say according to a script created by producers. Even some scenes that look real might be set up just for the cameras.

However, “Deep Fake Love,” brings another layer of fake to reality TV – AI-generated content designed to unsettle, bewilder, and manipulate the participants of the show.

Using deepfake to manipulate emotions

The show begins with five couples who arrive to demonstrate their love and loyalty to each other while being filmed. These couples are split into two separate villas, where they encounter attractive single men and women. This setup is intended to challenge their commitment to each other and their power of will.

Both villas throw parties day and night. The atmosphere fills up with sexual tension, and the participants become entangled in temptations and lust. One night, caught by surprise, the participants are invited into the studio, where the presenter shows them videos of their significant other being engaged in not-that-holy interactions with other men or women. Tears and fury follow the disclosure of the videos. The participants can’t believe their loved ones cheated on them at the very beginning of the show, where they actually came to prove their love to each other.

However, as the drama builds up, the presenter discloses a surprising twist – not all the videos were authentic. Some of them were created using doubling actors and deepfake technology to superimpose participants’ faces onto the video footage.

The game begins. Not all the videos are fake, and some couples really did cheat, were caught on camera casting doubts about their relationships, or were talking negatively about each other. But can’t know what’s real and what’s faked.

From this point on, participants have to participate in these weekly emotional ‘tortures,’ watch videos of their loved ones kissing or shagging someone else, and decide whether the video is real or AI-generated.

The couple with the highest count of accurate answers will emerge as the winners of the show, securing the 100,000 euro prize. However, they’ll need to hold firm as it’s not only the money that’s at stake – but their relationship as well.

Deepfake technology

Evidently, as the storyline of “Deep Fake Love” shows, the entertainment industry is aligning itself with technological trends. Deepfakes began gaining traction on the internet in 2017, and the popularity of this technology has since sparked ongoing controversies.

Deepfake technology allows users to replace a face in a video with a computer-generated face that closely resembles another person. The programs that create deepfakes usually use two different AIs simultaneously.

The initial AI scans through the provided images, videos, and audio, crafting manipulated visuals. Meanwhile, the second AI assesses these creations against authentic content, flagging disparities. This iterative process continues until the second AI can no longer distinguish between fabricated and genuine elements.

Experts have labeled deepfakes as the most significant AI-driven threat to date. Fake videos have the potential to influence democratic processes and impact the dynamics of conflicts, altering the morale of opposing forces.

An illustrative example of such an instance was a fake video of President Zelensky encouraging Ukrainian people to lay down their arms, which spread across social media platforms.

AI-generated pornography

However, despite the potential harm that deepfake technology might cause to the establishment, its main threat is to the integrity of personal life. The biggest source of damage caused by AI-generated fake videos continues to be in the form of reputation attacks through defamatory, derogatory, and pornographic content.

Sensity, a company that identifies and monitors deepfakes, published a report in 2019 revealing that a staggering 96% of deepfakes were non-consensual sexual depictions, with 99% featuring women. The term “deepfake” itself originates in pornography. It was born on the Reddit platform, where members of a group called “deepfake'' used AI to put celebrities' faces on porn actors.

Deepfake imagery is profoundly troubling as it’s non-consensual, and managing its spread is proving to be extremely challenging. Approximately 1,000 deepfake videos are being uploaded to adult websites monthly, and videos featured on three major platforms – XVideos, Xnxx, and xHamster – have collectively garnered millions of views.

Apart from exploiting celebrities, the technology is also commonly used in revenge porn, where malicious former partners spread fake explicit videos of their ex-partners. Fake videos can also affect unsuspecting individuals. This year, a Twitch streamer called Brandon “Atrioc” Ewing was caught buying and watching deepfake porn of his female colleagues.

Does this form of entertainment cross the line?

One of the participants of “Deep Fake Love” sits on the ‘white chair of truth’ at the studio. “Is your relationship true love or false?” asks the presenter, and the participant needs to decide by pressing the red or blue button.

It’s hard to cold-mindedly decide after seeing so many fake and real videos of your loved one misbehaving without knowing which is which. Emotions reach boiling point and cause literal headaches, even to the spectators behind the screen.

The participant hits the red button – the love is fake. Everyone cries for the evident breakup. The couple have not passed the test. Sad, but is the show only testing the relationships of the couples participating? Or is it also testing the boundaries of our tolerance to the pervasive technologies that are still being discussed on ethical grounds? How much do the deepfake videos in the show differ from fake pornography? The videos are non-consensual (at least not on the record), and defaming the participants on purpose.

Though manipulating images is not a recent phenomenon, and has been practiced since the beginning of photography, the progress in technology has significantly simplified the process of accessing and using algorithms to modify videos. Our inherent human ability to distinguish between what’s fake and real is struggling to keep pace with rapidly advancing technology, leaving us in a vulnerable state.

A TV show that promotes the usage of deepfakes intervening in private lives and manipulating relationships for entertainment might be part of the bigger problem – the thrill of new technologies is more significant than the need to ensure the ethics of their usage.