e-flux #75, September 2016
→ download PDF
- German translation: “Anormale Enzephalisierung im Zeitalter des maschinellen Lernens”, in: A. Koppenburger et (eds) Versprechungen. Münster: Unrast, 2017. +PDF
- Spanish translation: “Encefalización anormal en la edad del aprendizaje de máquinas”, in: F. Maniglio et al. (eds) Capitalismo Financiero y Comunicación. Madrid: CIESPAL, 2017. +PDF
- Czech translation: “Neobvyklá encefalizace ve věku strojového učení”, Big Light, Brno: Brno House Of Arts, 2016. +PDF
- Portuguese translation forthcoming: “Encefalização anormal na era da aprendizagem das máquinas”.
Even if artificial intelligence never awakes one day as a sentient being, there are already millions of machine learning algorithms that day-to-day scavenge gigantic data centers of social data to detect correlations, extract patterns, distil norms, predict tendencies, and make metadata mug shots of the population as a whole. Machine intelligence is not biomorphic—it will never be autonomous from humankind and, for sure, from the difficulties of capital, since it is a functional component of industrial planning, marketing strategies, securitarian apparatuses, and finance. Machine intelligence is sociomorphic, but not in a good way. Machine intelligence mirrors social intelligence in order to control the latter. The Turing universe is one of those magnifying mirrors, and it makes the collective body look grotesque, disproportioned, abnormalized by the glitches of computational power. We feed algorithms our racist, sexist, and classist biases and in turn they warp them further. As Marx knew, absent political action, machines do not just replace but amplify previous divisions of labor and social relations. Turing machines are no exception: datasets of populations educated in fundamentalism project an even more fundamentalist machine intelligence pattern.