Makysm Andriushchenko

Short Bio

Maksym Andriushchenko is a Principal Investigator at the ELLIS Institute Tübingen and the Max Planck Institute for Intelligent Systems, where he leads the AI Safety and Alignment group. His research develops algorithmic approaches to reduce the risks of advanced general-purpose AI, with a focus on aligning autonomous LLM agents, rigorous AI evaluation, model generalization, and frontier-model safety. He is a founding advisor at Exponential Security Labs, a member of the EU AI Act Scientific Panel, and a chapter lead for the International AI Safety Report 2026. His benchmarks, including JailbreakBench, AgentHarm, and PostTrainBench, are used by leading AI companies and safety institutes to evaluate frontier models.