The chief constituent of “Machine Learning” (2020) is a panel encrusted with black volcanic sand, a material widely available in Bali, where Sciascia is based. Set into the tactile, granular surface of the panel is a diagram that maps out the components of machine learning, composed of hundreds of letters and lines cast from fossilized resin. Machine learning is a form of artificial intelligence, a vernacular term for algorithms that are premised on the use of statistics to track patterns in large amounts of data, and thus able to analyse and predict user behaviour; models of machine learning are utilized by search engines and social media feeds, as well as recommendation programs on streaming services. Here, medium and motif collide: the immateriality of advanced computer systems – of mechanical cognition – is belied by a substance that provides the material basis of computer hardware. Sand, of course, is one of the sources of silicon, mined from the earth as silica sand or quartz, and used in the making of technological hardware such as microchips, memory cards and other components of integrated circuits. Sciascia is interested in the biological starting points of contemporary technology, and its origins in natural materials that were often formed millennia ago – the intersection of the digital and the mineral, the relentlessly contemporary and the occluded primitive.