Agents, Unleashed. Governance, Built In.

Nearly all teams are building agents. The hard part is trusting what they do.

Every agent that searches the live web, calls a tool or answers a client creates questions your CISO, risk team and regulators will ask:

• Where did that answer come from, and can we prove it?

• Which sources, tools and MCP servers is the agent allowed to touch?

• How do we give teams real AI leverage without opening a shadow-AI problem?

The teams pulling ahead have stopped treating governance as a brake. They treat it as the thing that lets them ship.

This roundtable is about how they do it — no slides, no pitches, just practitioners comparing notes.

What we’ll dig into

Practical takeaways, not theory.

1. From pilot to production. Which agent use cases are actually live at your peers’ companies, and what got them past security and compliance.

2. Autonomy vs. control. Deep search agents, sub-agents and coding agents now run multi-step work on their own. How do you let them run while still controlling what they access, what they do and what they deliver?

3. Trusting what agents find. When an agent researches the live web, how do you know its sources are real, current and approved, and can you show your work later?

4. Guardrails on what agents can touch. Which tools, data and MCP servers agents are allowed to use, and how to enforce that while they run.

5. Taming shadow AI. Giving teams approved, powerful tools so they stop reaching for unapproved ones.

6. Who owns AI risk? How security, AI enablement, platform and business leaders split the work without slowing each other down.

Note: this event is designed for senior leaders, founders, and industry experts