1Password and Apono reset access governance for humans, machines, and agents

At a glance

What this is: 1Password's acquisition of Apono reframes identity security around runtime access governance, using just-in-time controls for humans, machines and AI agents to replace standing privilege.

Why it matters: It matters because IAM, IGA and PAM teams now have to govern what identities can do at execution time, not just who can authenticate or where credentials are stored.

Context

The access governance problem is no longer limited to authentication or vaulting. Once humans, service accounts and AI agents can enter production systems, the control question becomes what they are allowed to do, for how long, and under whose authority.

That shift is especially important for non-human identity programmes because AI agents and workloads do not fit cleanly into human-era access review and approval models. If access is granted once and left in place, the governance layer loses sight of the action layer.

1Password frames the issue as a fragmented control plane problem: credentials are secured in one place while runtime permissions are managed somewhere else. The article argues that this split leaves standing access in place and makes policy enforcement too late in the lifecycle.

Key questions

Q: What breaks when access is granted once and never removed for humans, machines and agents?

A: Standing access creates hidden privilege persistence, which means a task can end while the identity still retains the ability to act. That breaks least privilege, weakens attribution and leaves teams unable to prove that access stayed within the approved scope. The failure is operational, not theoretical: the longer access remains alive, the larger the blast radius becomes.

Q: Why do just-in-time controls reduce risk more effectively than pre-provisioned access?

A: Just-in-time controls reduce the time window in which a privileged identity can be misused. They also force access decisions to align with a real task and a current approver, rather than with an assumed future need. The result is less privilege accumulation, less standing exposure and a clearer record of why access existed at all.

Q: How do security teams know whether agent permissions are still valid?

A: They should look for expiry, logging and revocation tied to the original task. If the workflow has changed, the data boundary has expanded or the agent is still operating after the authorised window, the permission is no longer valid. The test is whether the control follows the mission, not whether the account still exists.

Q: What is the difference between credential security and access governance?

A: Credential security protects the secret or token itself, while access governance controls what that identity may actually do once authenticated. The distinction matters because a protected credential can still enable excessive or long-lived access if runtime permissions are not scoped, monitored and revoked. Mature programmes need both layers, not one or the other.

How it works in practice

Just-in-time access turns permissioning into a runtime decision

Just-in-time access governance means the system evaluates an access request at the moment it is needed, then creates the minimum account, role or permission required for that task. In this model, access is not pre-created as a standing entitlement. It is generated in the target platform's native permissioning system, monitored while in use, and removed when the task ends. That matters because the control point moves from provisioning time to execution time, which is where production risk actually emerges.

Practical implication: treat access issuance as a runtime control and require task-scoped revocation by default.

Intent-based access control adds a second policy check for agents

Intent-based access control is different from ordinary approval because it binds access to the delegated task's declared purpose and compares that intent with what the identity actually does. For AI agents, that creates a live control loop: access can narrow or be revoked if behaviour drifts from the stated intent. This is particularly relevant for agentic workflows because an agent may keep acting after the original business condition has changed, which means policy has to follow behaviour, not just identity.

Practical implication: define declared intent as a policy input and monitor for drift during execution.

Unified audit trails close the gap between access and action

A unified audit trail brings request, approval, grant, revocation and activity into one record so security teams can see who accessed what, when and why. That is more useful than separate logs for credentials and permissions because it links entitlement to action. In practice, this is what makes runtime access governance defensible for compliance, incident review and anomaly detection. Without that chain, teams can prove that access was granted but not whether the access stayed within the approved boundary.

Practical implication: correlate grant and action logs so reviews cover both entitlement and behaviour.

NHI Mgmt Group analysis

Standing access is now the core identity governance liability. The article makes the case that the problem is no longer secret storage alone but permissions that persist after the need has passed. That is a governance failure across humans, machines and agents because the control plane keeps authorising access after the business justification disappears. Practitioners should read this as a shift from vault-centric security to runtime privilege containment.

Identity review processes assume access lasts long enough to be reviewed. That assumption breaks when access is created on demand and removed as soon as the task finishes, especially for AI agents and short-lived workload actions. The implication is not just faster automation, but a different governance model in which the review target becomes issuance policy and intent drift rather than persistent entitlement. Access certification alone cannot see what no longer exists.

Human, machine and agent governance should converge on one policy surface. Fragmented controls create blind spots because each identity type is governed by a different tool, cadence or approval path. A unified access model is emerging because enterprises now need one control plane for credential custody, runtime authorisation and delegated agent behaviour. Practitioners should re-evaluate whether their current IAM, IGA and PAM boundaries still make sense when the same task may involve all three identity types.

Ephemeral access debt: The article surfaces a useful concept for the market, namely the hidden governance debt created when access is granted per task but not governed per lifecycle. Once organisations rely on dynamic access for agents and workloads, every exception, approval path and revocation delay becomes accumulated risk. The practical conclusion is that access governance must be measured by how quickly privilege disappears, not by how easily it was issued.

Agentic access cannot be governed as if the actor were passive. The piece points to behaviour-aware controls because AI agents can continue acting after context changes, making static privilege assumptions unreliable. That is a field-level signal that access governance is moving from identity states to action states, with accountability attached to intent, scope and runtime observation. Teams should expect their governance model to follow that direction.

From our research library:

• 59% of organisations say they lack viable alternatives to standing privileged access for NHIs and AI agents, according to Delinea research.
• Read next: Ultimate Guide to NHIs — Key Challenges and Risks

What this signals

Standing privilege is becoming the governance debt behind agentic adoption. When organisations let humans, workloads and AI agents share the same approval model, the access layer becomes harder to certify and easier to overextend. A runtime-first governance design is now the practical dividing line between controlled adoption and permission sprawl.

Ephemeral credential trust debt: short-lived access does not solve governance unless the issuance policy, approval context and revocation boundary are all visible in one control plane. For identity teams, the strategic question is no longer how to issue access faster, but how to prove that access disappeared when the work ended.

For practitioners

• Map standing privilege across humans, machines and agents Inventory where access is granted once and left in place, especially across cloud platforms, data systems and developer tools. Focus on the identities that can still reach critical systems after the original task is complete.
• Move approvals to the moment of need Replace pre-provisioned access with just-in-time issuance for time-bound tasks, and ensure revocation is automatic at task completion. Apply this first to production systems and high-impact data paths.
• Bind delegated agent access to declared intent Require agent requests to carry the business purpose they are allowed to execute, then compare that intent to the actions the agent actually takes. Narrow or revoke access when behaviour diverges from the declared task.
• Unify audit records for grant and activity Ensure request, approval, grant, revocation and action events land in one audit trail that can support anomaly detection and compliance review. Separate logs for credentials and permissions will miss the runtime story.

Key takeaways

• The article frames standing access, not credential storage alone, as the main governance problem for humans, machines and AI agents.
• The operational signal is a move toward task-scoped access, automatic revocation and unified auditing across identity types.
• Practitioners should reassess whether their current IAM, IGA and PAM boundaries can govern runtime behaviour, not just initial authentication.

Key terms

• Just-in-Time Access Request: Just-in-Time Access Request is a pattern that grants access only when it is needed and only for the duration required. It reduces standing privilege by making access temporary, policy driven, and task scoped. This approach is especially useful for contractors, sensitive systems, and short-lived operational work.
• Standing Privilege: Standing privilege is access that remains active even when no immediate task requires it. For NHI programmes, it is a common failure mode because long-lived credentials and persistent roles create unnecessary exposure. Reducing standing privilege usually means tighter expiry, on-demand access, and clearer review of who or what still needs access.
• Intent-Based Access Control: An access control model that evaluates not just what an agent is requesting, but the inferred intent and context behind the request, granting or denying access based on whether the action aligns with the agent's declared purpose.
• Unified audit trail: A single evidence stream that links identity proofing, authentication, consent, and executed actions across human and non-human actors. It matters because split logs prevent teams from reconstructing what the agent was allowed to do and what it actually did.

Deepen your knowledge

NHI governance, agentic AI identity, and machine identity security are core topics in our NHI Foundation Level course, the industry's only accredited NHI security programme. If you are building or maturing an identity security programme, it is worth exploring.