As Generative AI, LLMs and autonomous agents move into production, the enterprise attack surface is rapidly evolving. The challenge is no longer limited to AI policies or vendor assessments. Organisations must build continuous trust across the entire AI lifecycle.
In this session, David and Matthew explore what “code to runtime” means in practice, and how organisations can secure AI systems from development through to live operation. We’ll examine how to generate continuous evidence that models and workloads are behaving as intended, address the risks introduced by agentic workflows, and move beyond fragmented controls towards verifiable AI governance.
Topics include:
-The reality of AI governance
-Code to runtime security
-The evolution of AI threats
-The imperative human layer
Attendees will gain practical insights into managing AI risk while enabling innovation with confidence.