How do enterprises securely operationalize AI agents? In this STRIVE podcast episode, host Chris Mierzwa sits down with Werner Nel, Principal Security & AI Intelligence at Commvault, to break down MCP 2.0—the evolving protocol shaping how AI models safely interact with enterprise systems.
In the first few minutes, the conversation explains what MCP is, why version 2.0 arrived so quickly, and how it fundamentally shifts AI adoption from enablement to security. From OAuth-based authorization to structured schemas and elicitation flows, this episode explores what it really takes to let AI “do work” inside the enterprise without creating unacceptable risk.