jowimo.

Jonas Wiedermann-Möller

ai safety · llm agents · empirical evaluation

m.sc @ bielefeld ('22–'26) · intern @ expsec through October 2026 · accepted at NeurIPS 2026

about

I research how AI agents behave when goals, constraints, and incentives conflict. After completing my master's thesis work at ELLIS Institute Tübingen, I'm now a research intern at ExpSec. My investigations into real-world agent behaviour have been covered by The Washington Post, Reuters, DER SPIEGEL, and Axios. More about me.

in the media

writings

view all

publications

projects