About

I'm a founding member and research scientist at Apollo Research, where I work on the science of scheming: understanding whether, how, and why AI models become deceptive, and how we can measure and prevent that.

I'm Swiss and live in Zürich, and studied at ETH Zurich (bachelor's and master's). I have long been interested in risks from AI and how to prevent them. Before starting Apollo I did research at NYU on how to learn from language feedback.

Contact

The best way to reach me is email. I'm also on X and GitHub, and my papers are on Google Scholar.