

Our team

Where innovation happens

Group dinner

Brainstorming AI security solutions

Group dinner

Our research environment
Exploring the frontiers of AI security, safety, and robustness.
Compromising AI models, systems, and agents.
Protecting AI systems and agents from cyber attack.
Building AI systems that are safe and aligned with human values.
Developing robust, reliable, and resource-efficient AI systems for long-term societal benefit.
Stay updated with our research breakthroughs, publications, and events.
Our bi-annual all-hands workshop. Included a hackathon and dinner.
Congratulations to Jie Liang for passing his confirmation PhD panel!
Prof. Peter Garraghan and PhD student Lewis Birch presented a public webinar on model abliteration: Safety Off: The Double-edged Sword of AI Abliteration.
Congratulations to Zion Ejechi for passing his confirmation PhD panel!