Christian Schroeder de Witt

Christian Schroeder de Witt is Associate Professor of AI and Information Security at UCL, where he leads the Witt Lab. He founded the field of multi-agent security, which studies what can go wrong when AI agents interact with each other and with the world, and how to secure them. His research on covert collusion between AI agents and on hidden capabilities in AI models has been published at NeurIPS and ICLR and covered by WIRED and Quanta Magazine. He co-leads the OWASP Agentic AI Research Council, lead-organised the first multi-agent security workshop at NeurIPS 2023, and has published joint research with the UK AI Security Institute. He is an incoming EPSRC Open Fellow, a Royal Academy of Engineering Research Fellow, and a Schmidt Sciences AI2050 Early Career Fellow. Before joining UCL he was an Associate Professor at the University of Oxford, where he continues to supervise doctoral students.