Ubiquitous facial recognition expertise can expose people’ political orientation, as faces of liberals and conservatives constantly differ. A facial recognition algorithm was utilized to naturalistic photos of 1,085,795 people to foretell their political orientation by evaluating their similarity to faces of liberal and conservative others. Political orientation was accurately categorized in 72% of liberal–conservative face pairs, remarkably higher than probability (50%), human accuracy (55%), or one afforded by a 100-item persona questionnaire (66%). Accuracy was comparable throughout international locations (the U.S., Canada, and the UK), environments (Fb and relationship web sites), and when evaluating faces throughout samples. Accuracy remained excessive (69%) even when controlling for age, gender, and ethnicity. Given the widespread use of facial recognition, our findings have essential implications for the safety of privateness and civil liberties.
Right here is more from Michal Kosinki, revealed in Nature.