Conservative female politicians tend to look happier and more attractive in photos than liberal politicians, a new artificial intelligence study conducted in Denmark has found, Mandiner reports.
According to research published in Scientific Reports in March, artificial intelligence can tell a person's political affiliation with 61% accuracy by analyzing a single facial image.
The New York Post reported that scientists analyzed about 3,200 public photos of political candidates running in the 2017 Danish municipal elections using Microsoft's Azure Face API tool, thus measuring the person's emotional state.
According to the analysis, 80% of the faces showed a happy expression, while 19% could be read as neutral.
"For women, high attractiveness scores were found among those identified as conservative by artificial intelligence," the results read. “These results are credible given that previous research with human raters has also shown that
there is a relationship between attractiveness and conservatism
the article revealed.
Even more accurate results were obtained for men
For men, the results were even more accurate, at 65%, before the researchers removed visual images other than the man's face, such as shirt collars, from the photos.
Left-wing political men showed a more neutral, less happy face than their conservative counterparts, the study concluded.
Front page image: Illustration / Government Spokesperson Alexandra Szentkirályi / Source: Prime Minister's Cabinet Office
