The Economist and Guardian stories described a research paper in which Stanford University researchers Yilun Wang and Michal Kosinski trained a deep neural network to predict whether someone was straight or gay by looking at their photograph. Wang and Kosinski collected a set of training images from an Internet dating website, photos of nearly eight thousand men and nearly seven thousand women, evenly split between straight and gay. The researchers used standard computer vision techniques for processing the facial images. When given pictures of two people, one straight and the other gay, the algorithm did better than chance at guessing which was which. It also did better than humans charged with the same task.

We show that faces contain much more information about sexual orientation than can be perceived and interpreted by the human brain. We used deep neural networks to extract features from 35,326 facial images. These features were entered into a logistic regression aimed at classifying sexual orientation. Given a single facial image, a classifier could correctly distinguish between gay and heterosexual men in 81% of cases, and in 74% of cases for women. Human judges achieved much lower accuracy: 61% for men and 54% for women. The accuracy of the algorithm increased to 91% and 83%, respectively, given five facial images per person. Facial features employed by the classifier included both fixed (e.g., nose shape) and transient facial features (e.g., grooming style). Consistent with the prenatal hormone theory of sexual orientation, gay men and women tended to have gender-atypical facial morphology, expression, and grooming styles. Prediction models aimed at gender alone allowed for detecting gay males with 57% accuracy and gay females with 58% accuracy. Those findings advance our understanding of the origins of sexual orientation and the limits of human perception. Additionally, given that companies and governments are increasingly using computer vision algorithms to detect people’s intimate traits, our findings expose a threat to the privacy and safety of gay men and women.

All we really know is that a deep neural net can draw a distinction between self-selected photos of these two groups for reasons that we don’t really understand. Any number of factors could be involved in the variation of these facial shapes, ranging from grooming to attire to photo choice to lighting. At the very least, the authors would need to show a statistically significant difference between face shapes. They fail to do even this.