How to Secure MCP and Agentic AI with Teleport and AWS

Register Here: https://goteleport.com/securing-the-future-how-to-safeguard-mcp-and-agentic-ai-with-teleport-and-aws/?utm_source=nhimg

As enterprises adopt agentic AI and large language models (LLMs) to automate critical business processes and access sensitive data, traditional security strategies no longer keep up.

The Model Context Protocol (MCP), a new connector for AI systems such as Amazon Bedrock Agents is revolutionizing enterprise integration. But with this innovation comes risk: AI agents acting with autonomy expand the attack surface, and MCP connectors amplify exposure to identity misuse, credential leaks, and privilege abuse.

This session will show how Teleport’s Infrastructure Identity Platform and AWS can work together to build secure, scalable MCP-based architectures.

What You’ll Learn

Why Agentic AI and MCP Are Both Transformative and Vulnerable

• How MCP links LLMs and enterprise systems in real time.
• The unique risks of giving AI agents direct access to APIs, data stores, and workflows.
• Examples of real-world attack paths through unmanaged machine and AI identities.

Treating AI Agents as First-Class Identities

• Why AI systems must be governed like humans in Identity and Access Management (IAM).
• Eliminating static credentials (API keys, hardcoded tokens) and replacing them with short-lived certificates.
• Preventing privilege creep and identity sprawl across AI-driven workloads.

Building Secure AI Workflows with Teleport + AWS

• How to implement least privilege and just-in-time access for AI agents in AWS.
• Using Teleport’s fine-grained access controls to enforce policy-based permissions.
• Creating real-time audit trails of every AI action for compliance and security oversight.

Real-World Strategies for Security Leaders

• Designing ephemeral, compliance-ready deployments for MCP-integrated AI systems.
• Detecting anomalies in AI-agent behavior with continuous observability and session monitoring.
• Ensuring regulatory alignment (ISO 27001:2022, SOC 2, HIPAA, NIST) when AI agents handle sensitive workloads.

Who Should Attend

• Platform Engineers seeking to operationalize AI securely.
• DevSecOps Leaders driving Zero Trust and compliance initiatives.
• AI Architects building agentic AI into enterprise production environments.

Key Takeaway

Securing MCP and agentic AI isn’t just about adding another control—it’s about rethinking identity security for a new class of non-human identities. With Teleport and AWS, enterprises can confidently embrace AI-driven innovation while ensuring least privilege, visibility, and compliance at scale.