Across the world, experimental predictive algorithms are being deployed that promise to prevent welfare fraud. Now for the first time, an international team of journalists gained unprecedented access to one of these "suspicion machines" in the city of Rotterdam. An investigation by Lighthouse Reports, WIRED and Vers Beton demonstrates how governments all over Europe - and municipalities like Rotterdam - are deploying AI risk scoring systems in welfare fraud detection, revealing discrimination, wide-scale surveillance, and a trail of lives ruined. In this podcast investigative journalist Gabriel Geiger (Lighthouse Reports) shares what they uncovered and discusses it with media anthropologist Miriyam Aouragh (Westminster University London) and law Professor Sofia Ranchordas (Tilburg University). Examining Rotterdam’s welfare fraud algorithm they look at the serious flaws and biases it contains, the human impact caused by its decisions, and what it reveals about the future of AI.

Ask this episode

“What did the council decide on item 42?” — answered with the speaker and the moment.

Transcript

Related