Anna Hartz
Many people feel overwhelmed by how fast digital images have changed in the last few years. It used to be easy to trust a photograph. Now anyone can create a face that looks real with only a few clicks.
This has caused worry for everyday users who want to know what is genuine and what is made by a machine. It has also sparked debate about whether we can keep up with these changes. A new study from the United States gives a bit more clarity on the matter.
Object-recognition
Researchers from Vanderbilt University found that people with strong object-recognition skills are better at spotting faces created by artificial intelligence, according to WP. They say that perceptiveness matters more than general intelligence or experience with technology. They also say it matters more than natural skill at recognizing real faces. Their findings were published in the Journal of Experimental Psychology: General.
The research team, led by professor Isabel Gauthier along with Jason Chow and Rankin McGugin, created the AI Face Test. It is the first tool designed to measure a person’s ability to tell real faces apart from AI-generated ones. The team retested participants to be sure the results held up. People who had strong object-recognition skills continued to perform well each time. This suggested that perceptiveness is a stable trait and not a random advantage.
Improving the skill
Gauthier said in an interview that many media claims about AI tricking everyone are exaggerated. She explained that some people can tell the difference with ease. Others struggle more. Many fall somewhere in between.
The researchers also connected their findings to earlier work. In other studies, strong visual processing helped people detect nodules in chest X-rays. It helped with identifying cancer cells in blood samples. It helped musicians read notes and helped specialists judge gender from retinal images. This pattern shows that perceptiveness influences many tasks in daily life. It also shows that the same skill can help when facing new challenges created by AI.
The team believes that understanding who is more vulnerable to visual misinformation will become important. They suggest that improving object-recognition skills may one day help protect people from manipulation on social media and in everyday online spaces. They also believe the AI Face Test is only the beginning. As generative tools evolve, the need for perceptive skills will keep growing.
