How to Fight the Broligarchy by Democratizing AI

This talk is based on core ideas from my book, Democratizing AI (May 2026).

Calls for 'democratizing AI' have become ubiquitous during the generative AI wave. While prompting fears about existential threats, there's enthusiasm about people integrating AI into daily lives. Surely this will 'democratize AI' by making it accessible to ordinary citizens.

But that's misguided. Just because generative AI is at our fingertips, doesn't mean it's truly in our hands.

We face asymmetrical concentration of power, wealth, and information among a small group of largely US-based, increasingly antidemocratic AI deployers—the 'broligarchs'—while AI-related risks are widely dispersed.

This asymmetry creates a democratic legitimacy deficit requiring collective response. I draw from interdisciplinary work in political philosophy identifying 'democratic innovations'—novel mechanisms and processes to redesign decisions beyond voting or endless deliberation. I give concrete examples of what this could look like.

Dieser Programmpunkt wird durch die Stiftung Mercator unterstützt. / This programme session is supported by Stiftung Mercator.