TIME: 13:20 - 14:10
LOCATION: Boardroom 6&7 - Multi-vendor ecosystems

Your AI can answer a prompt, but what happens when it can take action? This session explores how to put AI agents to work while keeping control of their decisions, access and impact. Through a practical demonstration, we’ll show how clear instructions, shared context, limited permissions and approval checkpoints shape an agent’s behaviour. We’ll explore how to inspect its work, catch mistakes and stop it when necessary. Leave with a clear understanding of agentic AI and how to decide what it can do independently, what needs your approval and how to verify the result.