Humans Trust AI-Generated Faces More Than Real Ones, Claims Scary New Study
A new experiment showed how faces generated by artificial intelligence were indistinguishable from actual human faces. If that's not exciting (or scary) enough, the humans involved in the study also found AI-generated faces to be more trustworthy.
As deep fakes become more common on social networking portals, watchdogs and critics alike are urging specialists to take note of its dangers - especially in crime. The two researchers also urged caution against AI-synthesised text, audio, video, and image that are used to commit fraud and propaganda.
The study was published in Proceedings of the National Academy of Sciences USA and its co-author Hany Farid, a professor at the University of California, Berkeley was quoted by Scientific American as saying the following - "We found that not only are synthetic faces highly realistic, they are deemed more trustworthy than real faces... these faces could be highly effective when used for nefarious purposes."
What did the research entail?
Researchers' trained the AI with an array of images representing Black, East Asia, South Asian, and white faces of both men and women. They later compiled 400 real faces with complementary 400 synthetic versions.
Based on these models, they asked three groups - first comprising 315 people, second comprising 219 participants, and the third with 223 people to identify the fake from a selection of 128 images. The first group simply had to distinguish real images from fake ones while the second group received some training on how to spot fakes. The third group assessed the trustworthiness of fake AI-generated images on a scale of one (very untrustworthy) to seven (very trustworthy).
Also read:?Can Artificial Intelligence Be Ethical? What AI Thinks About Its Own Dangers
The results were truly shocking. The first group that received no training showed average accuracy of 48.2%. The second group (with training) was able to spot fakes with 59% accuracy. As for the third group meant to rate images, the synthetic AI-generated faces received a higher average rating of 4.82 compared with 4.48 for real people.
Also read:?Artificial Intelligence: 7 Achievements From 2021 Driven Entirely By AI
While this doesn't mean all AI-generated images of faces are at par with real faces, it does demonstrate how deep fakes may be easily mistaken for real images. What do you think about this groundbreaking enquiry into the prowess of artificial intelligence? Let us know in the comments below.?For more in the world of?technology?and?science, keep reading?Indiatimes.com.? ??
References
Willingham, E. (2022, February 14). Humans Find AI-Generated Faces More Trustworthy Than the Real Thing. Scientific American.?
