Securing your AI agents: A technical demo

This webinar is a follow-up demo to our previous webinar, "The Future of AI Security," where we discussed the evolving AI agent landscape, their security risks, and the essential controls needed to build AI agents securely.

In this session, we dive deeper with a technical demo showcasing how Okta can help you design secure GenAI applications that allow agents to act independently, while respecting user authentication, authorization, and consent.

In our previous webinar, we covered:

• Understanding AI Agents – What they are, how they work, and why they matter for businesses.
• Security Risks & Challenges – The top threats AI agents introduce.
• How to Secure AI Agents – Best practices and security controls to mitigate risks and build trust in AI-driven systems.

This demo builds upon those concepts, and you’ll learn how to:

• Authenticate users across social, enterprise, and custom identity providers.
• Call APIs on the user's behalf using security standards and securely managed tokens stored in a token vault.
• Enable asynchronous user authorization for sensitive or out-of-band actions.
• Enforce fine-grained access control to ensure agents only access authorized data and actions.

Speaker: Lucas Gomes, Senior Technical Marketing Manager, CIAM, Okta