It’s Like Tinder: Scientists Just Got AI To Read Minds And Create The Most Attractive Face

It sounds like an episode of Netflix’s ‘The One”— except this time instead of using DNA to find your perfect romantic match, artificial intelligence reads your mind and finds the face of your ideal mate.

But this isn’t Sci-Fi, it’s a real research study. Scientists just got AI to read people’s minds and determine the face they’ll find the most attractive. The possible applications is where it gets interesting. They range from insight into unconscious bias to targeted dating apps.

Researchers at the University of Helsinki and University of Copenhagen wanted to find out if an AI program could learn what features make a face attractive to an individual. And then they wanted to know if the computer could take it one step further and actually design new images of faces to a person’s idea of attractiveness.

“In our previous studies, we designed models that could identify and control simple portrait features, such as hair color and emotion. However, people largely agree on who is blond and who smiles. Attractiveness is a more challenging subject of study, as it is associated with cultural and psychological factors that likely play unconscious roles in our individual preferences. Indeed, we often find it very hard to explain what it is exactly that makes something, or someone, beautiful: Beauty is in the eye of the beholder,” says study author Michiel Spapé.

AI reading minds

The specific type of AI used is called a generative adversarial neural network (GAN), which first learned by creating hundreds of digital portraits. To investigate attractiveness, the study team hooked participants up to EEGs and had artificial intelligence monitor their brain signals while they looked at different faces. It worked. AI was able to take that information and create new faces that appealed directed to each study participant’s preferences.

“It worked a bit like the dating app Tinder: the participants ‘swiped right’ when coming across an attractive face. Here, however, they did not have to do anything but look at the images. We measured their immediate brain response to the images,” Spapé explains.

Machine-learning came next, as the GAN evaluated the data and began generating images of faces predicted to be attractive to the individual. The researchers then tested those imaged in a double-blind procedure against matched controls. The AI created images matched study participants’ preferences with an accuracy of over 80%.

And that is something new in the world of AI. “By bringing in brain responses to the mix, we show it is possible to detect and generate images based on psychological properties, like personal taste,” Spapé explains.

And that’s where things get interesting. If computers can understand personal preferences, which are hidden inside our brains, we might be able to unlock tough to tackle issues like unconscious biases.

“We may also be able to look into other cognitive functions such as perception and decision-making,” says Spapé. “Potentially, we might gear the device towards identifying stereotypes or implicit bias and better understand individual differences.”

And the researchers don’t seem to have considered how dating apps like Tinder could use this technology to filter the faces of potential matches.