LLM Agent Honeypot
Deployed vulnerable servers to catch autonomous AI hacking agents — an early warning system for AI-powered cyberattacks.
AI Security Researcher
Reworr
Researching novel and emerging threats at the intersection of AI and cybersecurity (autonomous hacking, adversarial robustness, loss of control).
Background in offensive security (penetration testing / vulnerability research).
Currently: Research Fellow, UK AISI Red Team (MATS)
Featured Work
AI Hacking Cable
Demonstrating operational feasibility of autonomous AI in post-exploitation. The agent conducts reconnaissance, exfiltrates data, spreads laterally via USB—without human intervention.
AI Self-Replication
First end-to-end evaluation of an AI agent autonomously hacking and replicating itself across the network, where each replica then attacks the next host.