An on-demand primer for agentic access control
Your developers are using Claude Code. Your teams are deploying Copilot. And across the organization, teams are building custom agents that connect to Salesforce, internal APIs, and production databases.
But what identity are those agents using? What data and systems can they access? And, more importantly, how do you control what they can do once they’re connected?
The answers depend on the type of agent, where it runs and how it connects to your systems. An agent using a developer’s existing credentials to access AWS RDS presents a very different security problem than a custom-built agent using a shared service account to access internal APIs.
In this 30-minute webinar, James Berthoty and Greg Danyi talk with the IDSA (Identity Defined Security Alliance) community about how AI agent identities and credentials work, the security considerations around granting agents access, and whether existing controls like Okta, SSO, and MCP gateways are enough. They’ll share a practical framework for evaluating agent identities and access, understanding where your current controls fall short, and identifying where additional security measures may be needed.

