The OxAI Safety and Governance team brings together undergraduates, postgraduates, and researchers to explore technical AI safety research aimed at reducing catastrophic risk from advanced AI systems.

Speaker series

We host talks and Q&As from AI safety researchers. Past speakers have joined us from DeepMind, Redwood Research, and the Future of Humanity Institute.

Intro to AI Alignment Course

The AGISF programme introduces the core concepts of AI safety, and open problems in the field, through seven weeks of curated readings and small group discussions.

Labs

Our Labs programme pairs students with experienced supervisors to conduct technical AI safety research projects. Previous research areas have included active learning and inductive bias.