Brief description
Thirteen London based drag queens are photographed for dataset that is used to train a neural network in order to create a deepfake drag cabaret experience. Besides 13 deepfakes traind on video stills of existing members of the drag cast the artist also created ‘Zizi’ trained with images of all of the performers. After being welcomed to the cabaret experience by "Zizi" the user can create a own deepfake drag cabaret experience picking characters and songs. However the deepfake illusion is fragile:" At times this breaks down when the A.I. tries to conceive impossible positions or combines multiple different queer identities; it can even reveal the skeleton tracking the deepfake is built on."