How can artists glitch gender through drag? Through this project called KINGGLITCH, I answer by using gender expression to speculatively glitch surveillance systems through moving image and code. Facial recognition technology is all around us: in anti-theft software in supermarkets, airport terminals, some CCTVs, and more. These systems can contain Automatic Gender Recognition (AGR) algorithms, which typically sort bodies into binary genders. This work responds by combining Queer theory, surveillance studies and Glitch feminism to disrupt these systems through the use of putting on Drag King makeup. A glitch is a technical mistake, which means the glitch is a disruption. It troubles the idea of gender as a fixable, categorisable feature - instead the algorithm becomes uncertain, flickering between options and the grey areas. In this project, I trained a neural network on images of my face in and out of drag, to see where it fails and “glitches” in detecting gender. I then run this algorithm on a datamoshed/glitched video of me in zero, halfway, and full Drag King makeup as my king persona Chio Bussy. Through datamoshing, the moving image is also “dragged” visually, and the neural network attempts and fails to categorise a glitched and dragged image. A bounding box also displays where a facial detection algorithm “sees” a face in the glitched image, and we can observe how it both succeeds and fails unexpectedly - demonstrating how in some ways, the algorithms seem to work for us, but in other ways, they fail us. By putting this digital video on an analog CRTV, the signal is also “transed” through the literal translation of digital to analog signals. KINGGLITCH hopes to not just “gesture toward a politics of becoming-imperceptible” (Shabbar, 2018) through glitch, but become hyper-visbily-inperceptible through drag - by refusing being made invisible in the code, but so expansive that we are impossible to categorise.