AI agents in infrastructure: are your identity controls keeping up?

TL;DR: AI is moving from chat interfaces into code, APIs, customer workflows, and production systems, and Orca Security argues that once it can execute actions it must be governed like an operator, with defined permissions, human verification for sensitive actions, and auditable behavior. The identity lesson is that control boundaries built for passive tools break when the system starts acting inside the environment.

NHIMG editorial — based on content published by Orca Security: The Tool Becomes the Operator

Questions worth separating out

Q: How should security teams govern AI agents that can act inside production systems?

A: Security teams should govern them as operator identities with bounded authority, not as passive tools.

Q: Why do AI agents change traditional IAM assumptions?

A: They change IAM assumptions because identity is no longer only about authentication and access.

Q: What do security teams get wrong about prompt injection?

A: Teams often treat prompt injection as a content problem, when it is also an execution problem.

Practitioner guidance

• Classify AI systems by action authority Map each agent to the exact systems, APIs, and workflows it can affect.
• Gate sensitive actions with human verification Require a human checkpoint before any AI-driven change that can impact customers, infrastructure, or security settings.
• Instrument agent behaviour for auditability Log prompts, tool calls, outputs, and state-changing actions in a durable format that security teams can review after the fact.

What's in the full article

Orca Security's full research covers the operational detail this post intentionally leaves for the source:

• Examples of AI systems writing code, calling APIs, and taking customer-facing actions inside live environments.
• The article's framing of how prompt injection works when instructions and data share the same prompt space.
• Practical guidance on applying boundaries, human verification, and observability to AI operators in production.
• The security reasoning behind treating AI as an actor inside the environment rather than a passive assistant.

👉 Read Orca Security's analysis of AI agents as operators in infrastructure →

AI agents in infrastructure: are your identity controls keeping up?

Explore further

View Full Forum →  |  NHI Foundation Course →