Who owns AI? On democratization, control and power relations

Artificial intelligence is everywhere - in the world of work, in education, in our everyday decisions. And calls for the “democratization of AI” are getting louder and louder. But what does democratizing AI actually mean, in real-world terms? Access for all? Control by parliaments? Something else?

And should we even try to democratize AI in the first place, given that many people are not tech experts, and given that democracies often make bad decisions?

​It is easy to get distracted by the flashy disagreements currently driving public discourse on AI, from techno-utopian promises of salvation to dystopian warnings. But these claims risk distracting citizens from the most urgent problem in AI right here and right now: the problem that Big Tech oligarchs are unilaterally determining the direction of a powerful technology that is at all of our fingertips, but not truly in our hands. In a forthcoming book, Democratizing AI, political philosopher Annette Zimmermann draws on recent lessons from US and EU AI policy to defend the idea that democratizing AI is not only the right goal, but also one that is politically feasible in very concrete ways—if we take action now.

We talk about this with Milagros Miceli, sociologist and computer scientist, research lead at the Distributed AI Research Institute (DAIR) and head of the “Data, Algorithmic Systems and Ethics” research unit at the Weizenbaum Institute. Her internationally renowned work examines the working conditions and power relations in data work—in other words, the labor that makes AI innovation possible in the first place.

Thorsten Thiel, Professor of Democracy Promotion and Digital Policy at the University of Erfurt, will also be on the panel. He researches the effects of digital technologies on democratic processes and is one of the influential voices in the interdisciplinary discourse on digitalization and social participation.

Together, we will discuss how 'regular' citizens can take meaningful control of AI deployment and resist the misleading idea that innovation and democratic legitimacy are necessarily a zero-sum game.

The event will be held in English.