A recent European study conducted by Danish and Swedish researchers has found that conservative women are perceived as more attractive than their left-wing counterparts. The researchers used a deep-learning artificial intelligence known as a neural network to predict individuals’ political leanings based solely on their headshots. Interestingly, the study did not find any correlation between attractiveness and political leanings in men. However, it did reveal that left-leaning men tended to have more neutral and less happy facial expressions, suggesting their ability to better control their emotions.
While the researchers’ main objective was to demonstrate the accuracy of off-the-shelf AI in predicting political views based on limited information, they also highlighted the privacy concerns associated with this technology. The study showed that AI could correctly guess a person’s political leanings by analyzing a simple selfie posted on social media.
To conduct their research, the team selected 3,323 headshots of political candidates who participated in the 2017 Danish Municipal election. The neural network, capable of facial expression coding and classification, accurately predicted the political leanings of both men and women 61% of the time based on a single photo. The accuracy increased to 65% when all visual imagery, except the face, was removed from the dataset.
The researchers noted that the AI’s ability to predict political leanings was more successful with male candidates compared to female candidates. Additionally, the study found that physically attractive female candidates were more likely to be politically conservative, although this connection was not observed in the case of male candidates.
It is important to mention that the researchers eliminated 188 individuals from the dataset due to non-European ethnicity, as they believed this could impact the accuracy of the AI’s predictions. The AI utilized Microsoft’s Azure’s Cognitive Services Face API to assess the emotional expressions in the photos. The results showed that 80% of the faces displayed a happy expression, while 19% appeared neutral.
Although the study acknowledged certain limitations, such as biases in the photos presented by politicians and the AI’s difficulty in identifying certain facial expressions, the researchers were confident in the link found between attractiveness and conservatism among women. They emphasized that previous research conducted by human raters has also discovered a similar connection.
Interestingly, the neural network did not identify any prominent distinctions between right-wing and left-wing men based on typically masculine facial features or stereotypes. This suggests that such characteristics may not be reliable indicators of political leanings.
Overall, this study demonstrates the impressive capabilities of deep-learning AI in predicting individuals’ political views based solely on a photograph. However, it also raises concerns about privacy and the potential risks associated with this technology.
Denial of responsibility! VigourTimes is an automatic aggregator of Global media. In each content, the hyperlink to the primary source is specified. All trademarks belong to their rightful owners, and all materials to their authors. For any complaint, please reach us at – [email protected]. We will take necessary action within 24 hours.
