
Upcoming Broadcast
IS SSO enough to secure AI agents
December 9, 2026, 9:00am PT/12:00pm ET
IS SSO enough to secure AI agents
How AI agents authenticate and access enterprise systems using user credentials, service accounts, workload identities, and other forms of delegated authority.
Why SSO and authentication alone don’t determine what an agent should be allowed to do once it connects to a system.
How different agent architectures create different security risks depending on the identity, credentials, tools, and target systems involved.
Where controls like Okta, SSO, and MCP gateways help — and where gaps can remain across the full action chain.
A practical framework for evaluating agent access and identifying when additional runtime authorization and policy enforcement may be needed.
What will attendees learn?
- How AI agents authenticate and access enterprise systems using user credentials, service accounts, workload identities, and other forms of delegated authority.
- Why SSO and authentication alone don’t determine what an agent should be allowed to do once it connects to a system.
- How different agent architectures create different security risks depending on the identity, credentials, tools, and target systems involved.
- Where controls like Okta, SSO, and MCP gateways help — and where gaps can remain across the full action chain.
- A practical framework for evaluating agent access and identifying when additional runtime authorization and policy enforcement may be needed.
Who should attend?
- CISOs and security leaders
- Identity, IAM, and PAM leaders
- Cloud security and platform engineering teams
- AI and agent platform owners
- Security architects and engineers
- GRC and risk leaders evaluating controls for enterprise AI





