Derck Prinzhorn
AI Safety & Security
About
I'm an MSc AI student at the University of Amsterdam, currently doing my thesis on AI control at Max Planck Institute for Intelligent Systems, supervised by Maksym Andriushchenko.
My research focuses on AI safety and security, with particular emphasis on control, monitoring, and oversight of increasingly capable models. I also run Prinzhorn Solutions, where I help organizations understand and manage risks associated with adopting AI systems.
Previously, I worked as a Research Engineer at Aithos on pluralistic alignment, as AI Architect at the Dutch Police designing secure AI architectures, and co-founded Wisr (an EdTech startup). I graduated cum laude from my bachelor's, receiving the Amsterdam AI Thesis Award for my work on uncertainty quantification.
Experience
Max Planck Institute
Research on AI control under supervision of Maksym Andriushchenko.
Prinzhorn Solutions
Helping companies understand and manage risks associated with adopting AI systems.
Aithos
Worked on evals for AI value systems and moral competence.
Wisr
Worked on a startup helping teachers save time with AI grading.
Dutch Police
Defined reference architectures for AI, MLOps, and AI security.
Education
University of Amsterdam
Thesis on AI control at Max Planck Institute.
University of Amsterdam
Graduated cum laude. Amsterdam AI Thesis Award.
Publications
Highlights / News
January 2026
Started my thesis research on AI control at the Max Planck Institute for Intelligent Systems and ELLIS Institute Tübingen, supervised by Maksym Andriushchenko.
July 2025
Our paper HIVE: A Hyperbolic Interactive Visualization Explorer for Representation Learning was accepted at the Beyond Euclidean Workshop at ICCV 2025.
April 2025
Joined Aithos to work on pluralistic alignment and moral competence evals. Also founded Prinzhorn Solutions for AI consulting.
December 2024
Presented our reproduction of FairAC as a poster at NeurIPS 2024.
September 2024
Co-founded Wisr, an EdTech startup helping teachers save time with AI grading. Also visited the COPA conference in Milan to present our paper on conformal time series decomposition.
July 2024
Joined the Supervised Program for Alignment Research (SPAR), working on AI control with Aryan Bhatt from Redwood Research.
May 2024
Our reproducibility paper on FairAC accepted at TMLR. Our paper on conformal time series decomposition accepted at COPA.
November 2023
Received the AmsterdamAI thesis award for my bachelor thesis on uncertainty quantification in time series forecasting.
September 2023
Started my MSc in Artificial Intelligence at the University of Amsterdam.