Overview

Through training GANs (generative adversarial networks) on thousands of photographic portraits of pulled, intentionally marked and distorted faces, algorithms were lured into creating the unexpected. The result; a series of uncanny faces, familiar, yet distant. A homogenisation of the face, some adhering to and some transcending social and cultural norms. Many of the generated faces are unnerving, complete with lacerations and scarring, a result of original disrupted images going through the digitised learning process within virtual space.