A conversation with Kader Attia, artist and curator of the Berlin Biennale 2022, and Matteo Pasquinelli, professor in Media Philosophy, HfG Karlsruhe.

Date: Wednesday, 8 December 2021, 10:00-12:00
Place: online (link TBA)

Almost a decade after the rise of deep neural networks and machine learning (ca. 2012), it is time to draw up a balance of their cultural impact, to discuss what kind of knowledge model this new form of AI has come to embody and reinforce. Even without going into technical details, contemporary AI has already had an ideological impact by imposing the idea that a machine can perfectly imitate human skills such as “memory”, “intelligence”, “learning” and “perception.”

This process of anthropomorphisation and naturalisation of technology is however occurring also at a different level: notions such as “norm”, “error” and “anomaly”, for instance, are increasingly given mathematical definitions and uncritically translated into social ones. Meanwhile, we see that large repositories of knowledge and cultural heritage, from public libraries and museum collections to private photos and social media posts, are turned into training datasets for AI models, in a process of knowledge extractivism that renders all cultural heritage just as a data resource. At a deeper and more subtle level, the use of machine learning contributes to implicitly impose a new reductionist view to science as much as to culture, replacing the old episteme of causal explanations with an episteme of correlations

It is a new form of the colonisation of the mind, education, public institutions, social relations, collective memory and also nature that, this time, is not imposing a rationalist and mechanistic view of the world as in the modern age, but a statistical and algorithmic one for the purpose of economic profit (Shoshana Zuboff has in fact termed all of this surveillance capitalism). 

It is known that machine learning  can be beneficial as a new instrument in some applications, that its “intelligence” emerge from the imitation of the intelligence of our collective behaviours — we know this very well — but we should also not forget those fields of knowledge production that cannot or do not want to adapt to this model of knowledge and epistemology. What is the role of art, education and community practices, of alternative ways of thinking in the face of the growing hegemony of algorithmic thinking? Or to put it in another way: What happened to learning, to all the practices and institutions of education, in the age of machine learning?