Guest Lecture for COMP7029 at the University of Kent

On 3 June 2026, I delivered an in-person guest lecture at the University of Kent for the master’s module COMP7029: Artificial Intelligence and Cyber Security. The lecture took place at 9:00 AM in Sibson Seminar Room 7 and was attended by around 20 engaged master’s students.

I was invited by Dr Rogério de Lemos, Senior Lecturer and Director of Postgraduate Research (PGR) in the School of Computing.

The lecture, “From Chatbot to Cyber Agent: Governing AI that acts on cyber infrastructure,” explored how the security problem changes when AI systems move from simply answering questions to taking actions through tools. I introduced the idea of cyber agents, their opportunities and risks, and why governance is essential when AI systems interact with cyber infrastructure.

The lecture was based on my recent work, APIOT: Autonomous Vulnerability Management Across Bare-Metal Industrial OT Networks, which I used as the main case study to discuss autonomous vulnerability management in industrial OT networks. I also shared examples from the slides on how such systems can be evaluated across discovery, exploitation, patching, and verification tasks involving protocols such as CoAP, Modbus/TCP, and MQTT.

The session also included a short in-class exercise where students reflected on how to set boundaries for an AI agent scanning a small network of vulnerable IoT devices and recommending fixes. It was a pleasure to discuss this work with the COMP7029 students and see their engagement with questions around AI, cyber security, and responsible autonomy.

Slides from the lecture: From Chatbot to Cyber Agent (PDF).