Simon Storf

London

I work on AI Safety and Security at Anthropic. Before this I was a fellow in the Anthropic Fellows Program, working on red-teaming and AI control, and before that I worked with Marius Hobbhahn at MATS on monitoring for scheming. I studied computer science at ETH Zürich.

In a previous life I raced 60m and 110m hurdles.

Research

  1. Preprint, 2026

    Red-Teaming Auto Mode: Improving Blocking Classifiers Against Malign Coding Agents

    Alex Remedios, Simon Storf, Fabien Roger, John Hughes

  2. NeurIPS 2026

    Training Deliberative Monitors for Black-Box Scheming Detection

    Aditya Sinha, Akshat Naik, Victor Gillioz, Simon Storf, Kilian Merkelbach, Rich Barton-Cooper, Axel Højmark, Marius Hobbhahn

  3. ICML 2026

    Constitutional Black-Box Monitoring for Scheming in LLM Agents

    Simon Storf, Rich Barton-Cooper, James Peters-Gill, Marius Hobbhahn

* Equal contribution.