Friedemann Lipphardt

I’m a computer science researcher working on internet censorship, content moderation in large language models, and AI safety. I completed my PhD at the Max Planck Institute for Informatics under Anja Feldmann, and I’m now a Server Engineer at Mankind Games in Tokyo, building large-scale multi-tenant infrastructure for fleet surveillance, 3D modeling, and game backends, while shaping the company’s AI strategy.

My research has been published at NDSS, IEEE EuroS&P, FOCI and AINTEC. The throughline is measuring what large systems actually do to information — whether that system is a state-level VoIP censor in the Middle East, a frontier LLM moderating sensitive queries across 13 languages and 12 vantage points, or a TikTok recommendation feed targeting underage users. I’m a long-time contributor to oobabooga/text-generation-webui, the open-source LLM framework with 45k+ stars on GitHub.

Before Tokyo I taught Data Networks (~400 students/semester) and Hot Topics in Data Networks at Saarland University for four years, and I supervised an MSc thesis on LLM safety. I also did shorter research stints at IIJ Research Lab in Tokyo and at the University of Turku.

I speak English, German, Japanese, and French — feel free to reach out in any of them.