PhD Position AI Alignment: Value Assessment for Open Models & AI Systems

This role is a 5-year PhD position focused on advancing AI alignment for high-stakes law enforcement applications within the Netherlands Police. The candidate will establish ethical and legal value taxonomies for open-weights foundation models and engineer automated evaluation, red-teaming, and human-in-the-loop pipelines. This research aims to ensure AI systems operate safely and fairly while strictly complying with the EU AI Act and local operational requirements.