Teleport Introduces Beams Runtime for AI Agents With Identity

Teleport said it introduced Beams, a runtime for running Artificial Intelligence (AI) agents with identity and access controls. The company said Beams is designed to address the process of connecting identity, infrastructure, and auditing for agents in production environments.

Teleport described a challenge in agentic workflows for infrastructure: engineers want to develop, test, and deploy agents securely, but current agent launches require stitching together Identity Access Management (IAM), infrastructure, and secrets by hand. The company also said there is no consistent identity and no visibility into agent actions when teams build their own container or Virtual Machine (VM) workflows.

Beams runs each agent in an isolated Firecracker VM with built-in identity. Teleport said each Beam provides policy-controlled, tracked, and auditable access to trusted data sources, infrastructure and inference services without secrets or shared credentials. The company said Beams inherits delegated identity to authenticate to registered services and inference endpoints without using secrets, and that it supports fine-grained networking control and audited actions.

Teleport said Beams targets agentic use cases that include internal agents accessing production services, ephemeral workflows for developers building against staging without exposing secrets, and multi-agent production pipelines requiring hardened, reproducible isolation. “Security and IAM requirements interfere with getting AI agents into production. With Beams, security is a prerequisite that's already solved, so engineers can innovate freely knowing every agent has the identity, access controls, and audit trail it needs,” said Alexander Klizhentas, Teleport CTO. Beams will launch as an MVP on April 30, 2026.