Overview

Artificial intelligence is transforming work and raises fears of job loss, automation, and widening inequality. Yet the effects of new technologies are not predetermined: firms make choices about whether AI is used to replace or support workers. A growing literature argues that “pro-worker AI” is possible, but we know remarkably little about the institutional conditions required for it. Worker voice inside firms – who gets a say when new technologies are introduced – could be a potential lever for pro-worker AI, but remains largely unstudied.

This project studies how worker representation shapes firms’ AI adoption and use by focusing on German works councils: firm-level employee committees elected by the workforce, with legally mandated rights to be consulted on technology use, workplace organisation, and training. When firms introduce AI systems, works councils can audit proposals, negotiate safeguards, and push for reskilling and organisational adjustments. By combining administrative firm records with new employer surveys, the project provides some of the first evidence on how worker voice affects not just whether firms adopt AI, but what kind of AI they adopt and how it is used inside the workplace.

The project contributes to debates on the future of work by showing how firm-level governance institutions can shape technological change. Rather than treating AI as an exogenous force, it highlights worker representation as a potential policy lever to steer innovation toward productivity gains, skill enhancement, and job quality, rather than pure automation. In doing so, the project offers insights for policymakers and firms seeking to ensure that AI adoption benefits both workers and employers.