/Events /Agentic Safety & Security
About
AI agents can take action. How do we make sure they take the right action—with the right access, oversight, and safeguards?
Join the Global AI Community — Miami at Microsoft on October 20 for a two-hour meetup focused on Agentic Safety & Security.
As AI moves from answering questions to executing tasks, leaders and builders face new questions: What should an agent be allowed to do? What data should it access? When should a human step in? And how do we spot problems before they become business risks?
We’ll explore these questions together through practical discussion, shared experiences, and connections with others building and leading in AI.
Topics we’ll explore:
Agent permissions: Setting boundaries around tools, data, and actions.
Security risks: Understanding prompt injection, data exposure, and misuse.
- Human oversight: Knowing when to require approval, escalate, or stop an agent.
- Accountability: Making agent activity visible, traceable, and auditable.
- Responsible deployment: Balancing useful automation with safety and trust.