Agentic AI does more than generate answers. It takes actions across systems inside enterprise environments, and that raises new questions for security, governance and control.
Security leader and author Josh Woodruff joins Ronan Murray and Ian Finlayson to explain how agentic AI systems operate, why they should be treated like junior employees with system access, and how organisations can introduce them safely. Drawing on his book Agentic AI and Zero Trust, Josh sets out a practical framework for securing AI agents.
JW
Guest
Josh Woodruff
Security Leader & Author
Josh has decades of experience across enterprise security and co-wrote Agentic AI and Zero Trust, which introduces the Agentic Trust Framework for governing AI agents.
It amplifies everything. If your house is in order, it will amplify that. If your house is not in order, it will amplify the mess.
Josh Woodruff
In this episode
What agentic AI is and how it differs from generative AI
Real-world use cases across supply chain, logistics and operations
Why an AI agent should be treated like an intern with system access
The main risks, including prompt injection, data poisoning and rogue behaviour
Why kill switches and behavioural monitoring matter
How Zero Trust principles help secure autonomous AI systems
Highlights
Agentic AI is like a junior employee with system access.
Josh Woodruff
Trust needs to be earned, not granted.
Josh Woodruff
Attackers aren't breaking in anymore. They're logging in.
Josh Woodruff
The more constraints you put on, the better the security and the better the performance of the agentic system.
Episode 13