FaceSwap is the current, most accessible application for the creation of deepfakes available in online coding libraries. Its interface presents a grid of still frames taken from videos in the dataset that is fed to the machine, as well as a history of commands and prompts in coding language. It shows the learning process as it is visible to us, hiding beneath the surface the mathematical, non-human process of identifying how photorealistic image data is written and recreated. The truth is that we do not fully understand how machine learning systems arrive at their conclusions. The flatness, opaqueness, and inaccessibility of the process are the truths Interface wants to make sensible in regards to the technology, proposing a realism of the ontological fakeness of contemporary digital photographic processes.