AI Predicts Political Orientation Based on Facial Features: Stanford Study Raises Privacy Concerns

The use of Artificial Intelligence (AI) in predicting political orientation based on facial features has raised significant ethical concerns as the US gears up for the 2024 presidential election. A recent research paper published in ‘The American Psychologist’ by scholars from Stanford University revealed that AI can now predict political orientation with 70% accuracy by analyzing facial features.

The study, led by Associate Professor Michal Kosinski, tested the AI model on images of 591 individuals and found that conservatives tend to have broader, wider features, while liberals have narrower and smaller features. The model was also tested on 3400 celebrities and politicians with known political orientations, yielding similar accuracy levels.

This development has sparked concerns about privacy and the potential misuse of AI technology. Kosinski emphasized the need to protect individuals’ sensitive information, such as political orientation, from being exposed without consent. With the upcoming election, the use of AI for various activities, including generating realistic images and videos, has already been observed.

AI-generated videos criticizing political figures have circulated widely, often without proper context, contributing to misinformation. Kosinski warned that the study serves as a cautionary tale about the privacy threats posed by AI, particularly in the context of a high-stakes election. As the US prepares to elect its 47th president, the ethical implications of AI technology in politics remain a pressing issue.