Identity in the agentic era needs constant, omnipresent, adaptable control

At a glance

What this is: This blog argues that identity must become constant, omnipresent, and adaptable because AI agents expose the limits of human-era IAM, with delegated authority, sub-agent sprawl, and machine-speed execution outpacing quarterly governance.

Why it matters: IAM, IGA, PAM, and NHI teams need to rethink control timing, coverage, and identity taxonomy because agentic systems turn stale access review models into blind spots.

👉 Read C1.ai's analysis of identity control in the agentic era

Context

The core security problem is not that AI agents are new, but that they operate inside identity models built for humans. Human-era IAM assumed slow-changing access, bounded roles, and review cycles that could catch drift after the fact. AI agents change that pacing completely because they act continuously, delegate authority, and can reproduce into sub-agents faster than governance workflows can observe.

For IAM, NHI, and PAM programmes, the issue is broader than secrets or login friction. The article frames identity as the control plane for humans, agents, and delegated execution chains, which means access governance now has to cover machine-speed decision-making, omnipresent enforcement, and identity types that did not exist when most programmes were designed.

Key questions

Q: What breaks when organisations rely on periodic access reviews for AI systems?

A: Periodic access reviews break when the identity scope changes between review cycles. AI-enabled workflows can create, use, and retire access faster than reviewers can validate it, so certification no longer reflects reality. That leaves stale permissions active and makes breach exposure harder to detect before it is used.

Q: Why do AI agents increase non-human identity risk?

A: AI agents increase non-human identity risk because they can execute many actions quickly once they inherit a credential or tool permission. That speed expands blast radius, shortens attacker dwell time, and makes weak delegation more dangerous. The remedy is tighter scoping, continuous verification, and strict separation between observation and execution privileges.

Q: How do security teams know whether an agent identity is actually governed?

A: An agent identity is governed only when teams can identify the owner, locate the credentials, define the allowed scope, and revoke access without hunting across endpoints or backup files. If any of those pieces are missing, the identity is partially shadowed. The practical signal is whether access can be answered in minutes, not days.

Q: How should teams govern access when AI agents and service accounts share the same business systems?

A: Treat them as different identity subjects with the same governance obligation. Create one access model that covers ownership, entitlement scope, review cadence, and offboarding across human and non-human identities, then apply role-appropriate controls to each class. The goal is not separate programmes. It is one risk model that can follow access across systems and workflows.

Technical breakdown

Why quarterly access reviews fail for AI agents

Traditional identity governance assumes access persists long enough to be reviewed, recertified, and then adjusted. AI agents break that assumption because they authenticate repeatedly, change context continuously, and can complete millions of actions before a human review cycle notices drift. This is not just faster automation. It is a different control problem in which governance moves from post-event certification to real-time authorization and revocation. Event-triggered evaluation becomes necessary because calendar-driven governance cannot keep pace with delegated machine action.

Practical implication: move access governance closer to issuance and session enforcement instead of relying on periodic reviews.

How delegated authority turns into sub-agent permission inheritance

The article describes a delegation chain in which a human approves a parent agent, that agent spawns sub-agents, and those sub-agents inherit some portion of the original authority. Technically, the risk is not only overpermissioning but authority propagation without a matching accountability chain. Each inherited permission increases blast radius because the human sponsor no longer sees the downstream actor set. In identity terms, the subject of governance is no longer one principal but a reproducing graph of principals and inherited entitlements.

Practical implication: govern delegation chains as first-class identity paths rather than treating each spawned agent as an isolated workload.

Why continuous access evaluation becomes the control backbone

The post points to Continuous Access Evaluation Protocol as the model for always-on identity enforcement. The architectural shift is from static role assignment to continuous authorization decisions that can react to context, lifecycle state, and risk signals in near real time. This matters because agents may authenticate hundreds of times per minute and cannot wait for annual or quarterly governance. The identity system must therefore act like a live enforcement fabric, not a record-keeping layer. That is a structural change in how access is mediated across clouds and applications.

Practical implication: align identity architecture to continuous evaluation patterns where policy can react faster than the agent can move.

Breaches seen in the wild

• Scania insurance portal breach 2025: An attacker used an external user login, likely stolen by infostealer malware, to take insurance claim documents from a Scania portal.

Read and download The State of NHI & AI Agent Breach Report 2026, covering 150+ breaches impacting Non-Human Identities including AI Agents.

NHI Mgmt Group analysis

Access review processes were designed for access that lasts long enough to be reviewed. That assumption fails when the actor is an autonomous agent because privilege can be requested, used, and discarded within the same operating window. The implication is that governance built around quarterly certification is now structurally behind the behaviour it is supposed to control.

Delegation without visibility creates authority that reproduces. The article's agent-to-sub-agent chain shows why human approval of a parent principal is not enough when downstream actors inherit permissions outside the sponsor's line of sight. That is not simple sprawl. It is a governance model that loses track of who actually exercises authority.

Identity has become the control plane for machine-speed work, not a login utility. Once agents operate across clouds, applications, and jurisdictions, identity determines whether access can be constrained, revoked, and audited in time. NHI governance therefore has to treat agent identity as operational infrastructure, not a secondary directory problem.

Continuous enforcement is now a category requirement, not a maturity enhancement. The article is right to frame identity as constant, omnipresent, and adaptable because the old layering of periodic reviews, static roles, and after-the-fact detection no longer matches the execution tempo. Practitioners should read this as a programme design reset, not a tooling upgrade.

Identity taxonomy is now a moving target. Humans, service accounts, AI agents, and sub-agents do not fit neatly into one governance schema because they differ in behaviour, inheritance, and lifecycle. The implication is that IAM and NHI teams need composable control models that can absorb new entity types without redesigning the whole stack.

What this signals

Constant identity becomes a design requirement when agents act at machine speed. Security teams should stop treating identity governance as a periodic assurance activity and start treating it as an always-on enforcement layer. When access can be consumed faster than a review cycle can run, the control point has to move to issuance, session state, and immediate revocation.

Delegation chains are the new blast-radius problem. The practical question is no longer only who has access, but which downstream principals can inherit and extend that access without clear sponsor visibility. That makes ownership, offboarding, and revocation the critical programme controls for agentic environments.

For practitioners

• Define agent identity as a governed principal class Create a separate governance model for AI agents, including registration, ownership, delegated scope, and revocation responsibility, so agent activity is not hidden inside human account structures.
• Replace calendar-based review with continuous authorization Move from quarterly or annual certification to event-triggered access decisions that can react when an agent's context, task, or downstream delegation changes.
• Map delegation chains to downstream access paths Trace how human approval turns into parent-agent and sub-agent permissions across systems, then document where visibility ends and inherited authority begins.
• Design for zero standing privilege across humans and agents Require task-scoped access that is granted on demand and revoked automatically so dormant agent permissions do not persist after the work is complete.
• Test identity outage scenarios for agent shutdown Validate what happens when the identity service is unavailable, including whether agents continue operating, lose the ability to revoke access, or bypass governance controls.

Key takeaways

• AI agents expose a structural gap in identity programmes built around human review cycles, because access can be created and consumed before a governance process has any chance to intervene.
• The article's strongest evidence is the combination of machine-speed authentication, inherited authority, and sub-agent sprawl, which together make stale access controls look increasingly ceremonial.
• Practitioners need continuous authorization, explicit delegation governance, and zero standing privilege if they want identity controls to remain enforceable in agentic environments.

Key terms

• Agent Identity: An agent identity is the set of attributes, credentials and permissions assigned to an autonomous software entity. It is treated as a non-human identity because it can authenticate, act on systems and accumulate access over time, which creates governance, audit and lifecycle obligations similar to other production identities.
• Delegation Chain: A delegation chain is the sequence of identities, credentials, and tool calls an agent uses to complete a task across systems. It matters because each step may appear acceptable on its own while the combined path produces an outcome no reviewer would have approved directly.
• Continuous Access Evaluation: Continuous access evaluation is the practice of rechecking whether a principal should still have access after the session begins. In NHI environments, it matters because tokens and service accounts can remain valid while the surrounding risk changes, so enforcement has to follow the request, not just the login.
• Zero Standing Privilege: A control model in which an identity does not keep persistent access unless it is actively needed. For NHIs, this means credentials and permissions are issued for a narrow task and then removed. It reduces the time window and reuse value of stolen access.

What's in the full article

C1.ai's full blog covers the operational detail this post intentionally leaves for the source:

• The full delegation-chain argument showing how parent agents pass authority to sub-agents across systems
• The article's broader market context on why identity infrastructure must be constant, omnipresent, and adaptable
• Specific examples of standards and architectures such as continuous access evaluation and identity fabrics
• The vendor's framing of how platform consolidation is changing the identity control plane

👉 C1.ai's full post expands on continuous control, omnipresent identity, and adaptive governance for agents.

Deepen your knowledge

NHI governance, agentic AI identity, and machine identity lifecycle are core topics in our NHI Foundation Level course, the industry's only accredited NHI security programme. If you are responsible for identity security strategy or NHI governance in your organisation, it is worth exploring.