Pre-programme discovery projects
To help guide our thinking and shape the programme’s development, we funded a series of short, exploratory research projects running from December 2025 to February 2026. These projects range from exploring aspects of Arena design, diving into topics around physical trust and AI security theory, and running community events such as a hackathon.
Systemisation of Knowledge: Cryptographically verifiable supply chain data
Martin Klepmann, University of Cambridge
Trusted Molecular Memory: Exploring trust within hybrid computational systems
Larissa Pschetz, University of Edinburgh
Living Proof
Kimia Witte, University of Strathclyde
Formally verified coordination for trustworthy multi-agent systems
Kiran Gopinathan, Basis AI
Adversarial games for trustworthy AI systems
Davide Crapis, Independent
CME tournament
Enrico Botazzi, Independent
Untrustworthy Things Everywhere: An agentic challenge for securing cyber-physical industrial infrastructures
Awais Rashid, Hacktronics
LabTrust-Gym
Mateo Petel, Stanford University
Who am I? Digital identity in embodied AI
Mary Maller, Inversed Tech
SoTA’s cyber-physical security community
Jamie Croucher, Society of Technological Advancements (SoTA)
Benchmarking automated mechanism design
Samuele Marro, Institute of Decentralised AI
A substrate-independent, diagrammatic foundation for cryptographic trust
Martti Karvonen, UCL
AgentMatrix: Towards automated benchmarking and securing of AI agent interactions
Mario Kogias, Imperial College London
Physical-Layer Trust: Measuring embodied interaction in distributed robotic systems
Suet Lee, Autodiscovery Ltd.
Trust Everything, Everywhere
In the Trust Everything, Everywhere opportunity space we believe new trust infrastructure that extends formal security reasoning into the physical world will unlock trillion-pound cyber-physical markets and a society free to trust everything, everywhere.
Scaling Trust
The Scaling Trust programme seeks to create the capability for AI agents to securely coordinate, negotiate, and verify with one another on our behalf.