This AI algorithm predicts your first impressions based on your facial expressions

• Researchers from Chicago taught AI to model first impressions of humans.
• The AI uses thousands of computer generated face images to predict impressions.
• Some of the findings include parameters such as trustworthiness, intelligence, and electability, among others.

Researchers from Stevens Institute of Technology in collaboration with University of Chicago and Princeton University have taught an AI algorithm to model first impressions and predict how people will be perceived based on photographs of their faces. The study was published in the journal ‘Proceedings of the National Academy of Sciences’.

Jordan W. Suchow, a cognitive scientist and AI expert at School of Business at Stevens said “there's a wide body of research that focuses on modelling the physical appearance of people’s faces.” He also added, we are bringing this together with human judgments and using machine learning to study people’s biased first impressions of one another.

During the research, Joshua Peterson, Thomas Griffiths, Stefan Uddenberg, Alex Todorov and their team in Chicago asked thousands of people to give their first impressions of over 1000 computer generated images of faces, using criteria like religious, intelligence, electability, and trustworthiness. The recorded response was used to train a neural network to make judgments about people based on the images of their faces.

Jordan W. Suchow said “given a photo of your face we can use an algorithm to predict what people’s first impressions of you would be and which stereotypes they would project onto you while looking at your face.”

People who smile tend to be more trustworthy
Suchow also explained the working of algorithms, the findings are aligned with common intuitions and cultural assumptions. People who smile more tend to be more trustworthy and people with glasses tend to be more intelligent. According to him, the algorithm doesn’t provide targeted feedback.

Originally, the algorithm was developed to help psychologists in social experiments on perception.

The new algorithm can be useful in the real world, too. For instance, people carefully curate their public presence. Sharing photos makes them think they look intelligent, confident, and attractive. The algorithm can be used to support this process, said Suchow.

On the other hand, the algorithm can be also used to manipulate photos to make their subject appear in a specific way. Political candidates can appear more trustworthy by using this algorithm. Other AI tools are already being used to create deepfake videos.

To safeguard this technology, researchers have secured a patent and now they are creating a startup to license the algorithm for ethical purposes. Suchow said, “we are taking all these small steps so that no harm is done while using this algorithm.”

SEE ALSO:
How to choose the right gaming monitor
Google rolls out new app privacy labels in Play Store with better transparency for users