It is becoming increasingly difficult to determine whether a face is real or generated by artificial intelligence. The University of New South Wales (UNSW) has recently launched an AI faces test that challenges users' ability to distinguish between authentic and AI-created portraits.
The Test Experience
Guardian Australia's Carly Earl, a professional photographer, and Matilda Boseley, a self-proclaimed internet addict, took the test to see if they could rely on their expertise or intuition. The test presents a series of faces, some real and some AI-generated, and asks participants to identify which is which.
As AI technology advances, the line between reality and fabrication blurs. The test aims to raise awareness about the sophistication of AI-generated images and the potential for misuse. For photographers and everyday internet users alike, distinguishing real from fake is no longer a simple task.
Science or Vibes?
Earl and Boseley approached the test with different strategies. Earl looked for subtle cues in lighting and texture, relying on her professional training. Boseley, on the other hand, trusted her gut feeling, shaped by years of consuming online content. Their results highlight the growing challenge posed by AI in visual media.
The UNSW test is part of broader research into AI's impact on society. As AI-generated content becomes more prevalent, such tests help educate the public about the need for critical evaluation of digital images.
