Christian Schroeder de Witt DPhil, MSc Comp Sci, MPhys (Oxon)

Christian Schroeder de Witt is Principal Investigator of the Oxford Witt Lab (OWL, wittlab.ai) at the Department of Engineering Science, University of Oxford. He is an incoming EPSRC Open Fellow, and currently also holds a Royal Academy of Engineering Research Fellowship and a Schmidt AI2050 Early Career Fellowship, and has secured over £3.5 million in funding as sole PI. His research focuses on the foundations of trustworthy AI systems, with particular emphasis on multi-agent security - a field he recently defined to address system-level risks in interacting AI agents. His work introduces fundamental limits of detectability and motivates security-by-design approaches for agentic systems, with recent contributions on collusion, illusory attacks, and backdoors published at NeurIPS and ICLR. Christian’s research spans theory and practice, bridging AI, security, and complex systems, and includes collaborations with partners such as LLNL, BBC Verify, Adobe, and Google DeepMind. Prior work includes a widely recognised breakthrough in perfectly secure steganography and foundational contributions to deep multi-agent reinforcement learning.

OWL’s mission is to de-risk the large-scale deployment of advanced AI by building trust through technical assurance. Technical assurance is the disciplined, auditable evidence that an AI system will behave within specified safety and security bounds - by design, under test, and in operation - even in adversarial and multi-agent settings. We work from first principles to practice: from the mathematical foundations of multi-agent security and undetectable threats to next-generation evaluations and real-world demonstrations in domains such as open-source intelligence, biology, and climate science.

We are goal-driven and method-agnostic. Depending on the question, we draw on high-dimensional anomaly detection, game theory, RL/MARL, cryptography and steganography, and interpretability (theory and practice). We publish across leading AI, security, and interdisciplinary venues.

Systemic Assurance Foundations

OWL is pioneering the field of multi-agent security, addressing a critical gap in current AI safety and security by unifying perspectives from game theory, cryptography, and the complexity sciences into a coherent threat taxonomy. Our work on perfectly secure steganography, secret collusion, illusory attacks, and unelicitable backdoors, and other practically or theoretically undetectable threats establishes an information-theoretic basis for security-by-design in the AI age.

Alongside architectural guarantees and secure-by-design approaches, we develop principled anomaly detection methods tailored to low-signal regimes and out-of-distribution dynamics relevant to deployed systems.

Benchmarks and Evaluations

Robust assurance needs evidence. OWL designs next-generation benchmarks, red-teaming playbooks, and evaluation protocols for multi-agent and mixed-autonomy settings. We periodically assess the utility of alternative detection and mitigation tools - including white-box methods informed by mechanistic interpretability, unlearning, model editing, jailbreak defences, AI debate, watermarking, and alignment techniques - explicitly measuring limits, failure modes, and deployment value.

Capabilities and Real-World Applications

Assurance cannot be separated from how capabilities related to assurance evolve. We build select, high-signal capabilities - for example in multi-agent reasoning and long-horizon reasoning, coordination, and verifiable code generation - to better characterise risk, to stress-test assurance tools, and to inform standards. We translate results into real-world applications and collaborations across government, industry, and civil society.

Teaching

I have teaching experience at both undergraduate and doctoral levels within Oxford. As a Stipendiary Lecturer in Computer Science at St Catherine’s College, I delivered small-group tutorials in Machine Learning and core topics, teaching cohorts across multiple colleges and receiving consistently very positive student feedback. I have also contributed to doctoral training through the AIMS CDT at Oxford, where I developed and delivered advanced material in AI-related areas, combining theoretical foundations with interactive components for PhD students.

DPhil, MSc/MEng, and Undergraduate Supervision

If you are interested in working with OWL, whether as PhD/DPhil student, visiting student, MSc/MEng, Part B/C student, please get in touch. Some example directions can be found here: https://wittlab.ai/student_projects/