A.I. can read attractiveness from your brainwaves
Maybe beauty lies in the eye of the beholder but by further analysis, it resides in our brainwaves.
In our mind lies the concept of beauty, like many other personal thoughts. Although our private inner thought is something we’ve always considered inaccessible doesn’t mean they can’t be monitored and perhaps predicted by an AI.
Researchers used electroencephalography (EEG) measurements to detect what kind of facial features people find attractive, then they fed an artificial intelligence with the results so that the machine learning algorithm (a GAN aka a Generative Adversarial Network), to make it learn what kind of faces people found desirable, in order to create new ones designed specifically to please them.
The experiment was run at the University of Helsinki (Finland) by a team of psychologists and computer scientists to 30 volunteers. Participants, wearing elastic caps with electrodes, were sat in front of a screen and were shown a series of realistic-looking faces generated by a dataset of about 200,000 images in order to have their brain activity measured as they were looking at the faces without telling anything.
“They did not have to do anything but look at the images”, explained the cognitive neuroscientist Michiel Spapé. “We measured their immediate brain response to the images”.
The GAN was responsible to interpret brain responses according to the neural activity in order to learn whether a face was found attractive by the viewer. Basing on that data, the GAN was then able to generate new attractive faces.
In a second experiment instead, the newly generated faces were displayed back, alongside other randomly generated faces to the volunteers who were asked to rate their attractiveness.
The results validated the researchers’ test with volunteers rating fake attractive images as attractive for 80%, while the others were only chosen for 20%.
Anyway, this is just another example of how AI algorithms are getting ever more able to understand the world, even our most private things.
“Succeeding in assessing attractiveness is especially significant, as this is such a poignant, psychological property of the stimuli”, Spapé said.
“If this is possible in something that is as personal and subjective as attractiveness, we may also be able to look into other cognitive functions such as perception and decision-making. Potentially, we might gear the device towards identifying stereotypes or implicit bias and better understand individual differences”.
However, this is also scary. We have always thought of our mind as something inscrutable but if an AI could know us so deeply, it would be dangerous. Not because is good to lie but because having something that belongs exclusively to us, like our mind, can save us from the deceits of the rest of the world: it’s our safe zone. Being mentally naked means being without shrewdness or cleverness since everything would be out in the open, and we could be nothing but a victim as if anyone could anticipate our moves, and we’re going to lose that part of us which is only ours.