According to a recent study, Artificial Intelligence can predict a person’s political ideology with 61% accuracy by analyzing just a photo of their face. The researchers used approximately 3,200 public photos of political candidates who participated in the 2017 Danish municipal elections and evaluated them in terms of their emotional state using the Microsoft Azure Face API tool. These findings suggest that facial features can provide clues to a person’s political affiliation and highlight the potential of Artificial Intelligence in political analysis.
In the analysis carried out, it was observed that 80% of the faces presented a happy expression, while 19% considered themselves neutral. In addition, it was discovered that Artificial Intelligence can predict a person’s political ideology with 61% accuracy by analyzing only a photo of their face. The results were even more accurate for men, with 65% accuracy, before the researchers removed the extra visual imagery., such as shirt collars, from photos of men. These findings highlight the ability of Artificial Intelligence to identify certain facial patterns associated with political ideology and its potential application in political analysis studies.
For women (though less so for men), high attractiveness scores were found among those the model identified as likely to be conservative.
“For women (although not so much for men), high attractiveness scores were found among those the model identified as likely to be conservative”, reads the report. “These results are credible given that previous research with human raters has also highlighted a link between attractiveness and conservatism,” the study states.
Left-leaning male politicians showed more neutral and less cheerful faces than their conservative counterparts, the study details.
We also found that expressing happiness is associated with conservatism for both genders.