Me and my robot draw together. We map chess moves, draw glyphs and stick figures, sketch the paths of systemic forces. In this way, new asemic scripts are created. New languages. New cartographies. Although they may seem devoid of meaning, like mere visual play, such scripts offer a new grammar for viewing and feeling the structures in which we live today.

‍

This work illustrates how much human effort is needed for an artificial neural network to learn to draw simple human-like figures. The algorithm is trained on a database of 31,300 human-like glyphs – which I hand-drew for weeks and then digitally varied – and based on them, a statistical model of the distribution of points and lines is formed. This creates a program that draws new stick figures, a new pictographic language.