TL;DR: Oak outlines machine identity governance that only works when each identity has a named owner, approved scope, usage-based review and source-system retirement; the article ties control failure to orphaned identities, standing privilege and credentials that outlive the workload. Governance fails when records exist without evidence that access was actually issued, used and removed.
At a glance
What this is: NHI Mgmt Group's analysis, drawing on Oak's blog post: Machine identity governance depends on lifecycle evidence, and the control fails when ownership, approval, usage review and source-system retirement do not leave proof.
Why it matters: IAM and security teams need machine identity controls that prove access was justified, used and removed, because governance records without operational evidence leave standing access in place.
By the numbers:
- The 2026 Identity Security Landscape survey of 2,930 security decision-makers found 109 machine identities per human, and 40% of machine identities had access to organizational data.
👉 Read Oak's guide to machine identity governance from creation to retirement
Context
Machine identity governance fails when organisations treat service accounts, API keys, tokens and certificates as a static inventory instead of a lifecycle with evidence at every stage. When no joiner-mover-leaver process covers these identities, the records drift away from the systems that actually authenticate and authorise them.
The governing question is not whether an identity exists, but whether it still has a valid owner, a current purpose and a recorded reason to keep access. That is why lifecycle governance for machine identities has to connect issuance, usage, rotation and retirement in one chain, not in separate tickets.
A comprehensive NHI reference helps anchor the lifecycle view, but the operational test remains whether the target system still recognises the identity after governance says it is closed.
Key questions
Q: How should machine identity governance be handled from creation to retirement?
A: Machine identity governance should follow the full lifecycle: assign an owner and purpose at creation, approve scope before credentials are issued, review actual usage, rotate credentials by risk and retire the identity in the source system. The central control is evidence, because a record that does not change the target system leaves access alive after the governance record closes.
Q: Why do service accounts become risky when no owner is assigned?
A: A service account without an owner becomes risky because nobody is clearly accountable for its scope, secrets, rotation, or retirement. Orphaned accounts often keep broad permissions, shared credentials, and long-lived tokens after the application changes, which makes access review, incident response, and offboarding difficult. Assign a named owner, record the business purpose, and tie it to periodic attestation and secret rotation.
Q: How do you review machine identity access against actual use?
A: Review machine identity access by comparing granted permissions with observed activity, such as API calls, job schedules, authentication logs, and secret usage. Remove entitlements that are never used, narrow broad roles, and validate whether the service account, token, or certificate still matches the workload it serves. The goal is to detect standing access that no longer reflects actual runtime need.
Q: What is the best way to retire a machine identity safely?
A: Retire a machine identity by first identifying every workload, integration, secret, certificate, and scheduled job that depends on it, then replacing it with a new identity where needed. Revoke tokens and keys, disable the account, remove trust relationships, and confirm no live process still authenticates with it. Secure retirement ends with verified decommissioning and ownership records updated.
Technical breakdown
Why inventory is not governance for machine identities
An inventory tells you what was discovered at a point in time. Governance tells you who owns the identity, why it exists, where it acts and when its access should end. Machine identities often come from pipelines and scripts, so they bypass human joiner-mover-leaver workflows and never gain the lifecycle context that lets reviewers judge whether the identity is still legitimate. The result is orphaned access that looks controlled on paper but remains active in the target system. See the NHI Ownership and Accountability Guide.
Practical implication: Treat discovery as the starting point for ownership and purpose assignment, not as proof that the identity is governed.
How approval, scope and issuance should line up
Approval only has control value if it happens before credentials exist and if the approver reviews the intended scope against the registered purpose. In machine identity governance, the request, the approval and the issuance event must form one auditable sequence. If a pipeline mints the credential first and the ticket is updated afterward, the control is procedural rather than preventive. Time-bound access and separate approver roles are what stop self-approval and after-the-fact justification. See the NHI Authentication Guide.
Practical implication: Require issuance to wait on a recorded approval so the target system never sees a credential without authorisation.
Why usage-based review beats periodic certification
A machine identity review is useful only when it compares granted access with actual activity. Usage data shows which entitlements remain necessary, which ones are never exercised and which ones have drifted beyond the approved scope. Periodic campaigns without usage context tend to rubber-stamp access because reviewers cannot see the difference between theoretical permission and live dependency. In lifecycle terms, the control is not review frequency alone, but review evidence that can drive removal decisions. See the IAM and IGA Basics.
Practical implication: Certify machine identity access against observed usage, then remove unused rights at the source.
Threat narrative
Attacker objective: The objective is to preserve reachable machine access that can be reused for lateral movement, data access or operational abuse after the original business purpose has disappeared.
- Entry occurs when a machine identity is created without a real owner, purpose or pre-issuance approval, which lets credentials exist outside governed lifecycle controls.
- Escalation follows when standing privilege or delayed rotation gives that identity more reach than its workload needs, and the access persists after the original need has ended.
- Impact appears when dormant or orphaned credentials still authenticate to target systems, leaving access live even after governance records say the identity is retired.
Breaches seen in the wild
- Dropbox Sign breach 2024: A compromised back-end service account gave attackers Dropbox Sign customer data, including API keys, OAuth tokens and MFA information.
- Microsoft Midnight Blizzard breach: Midnight Blizzard (APT29) exploited legacy test account without MFA to breach Microsoft.
Read and download The State of NHI & AI Agent Breach Report 2026, covering 200+ breaches impacting Non-Human Identities including AI Agents.
NHI Mgmt Group analysis
Machine identity governance is a lifecycle discipline, not a record-keeping exercise. Oak's framing is correct because the control only exists when ownership, approval, usage review and retirement all leave evidence in the target system. A catalogue of identities without source-system enforcement creates the illusion of control while access keeps working underneath it. Practitioners should measure governance by whether the identity actually stops authenticating when the record says it is closed.
Ownership is the first control, because everything else depends on it. When a machine identity has no accountable owner, rotation stalls, reviews are rubber-stamped and retirement never gets triggered. That is not an administrative nuisance, it is a structural failure in the governance model. The practical conclusion is simple: if ownership cannot be resolved, the identity is already out of policy.
Standing privilege is the failure mode that turns convenience into exposure. Machine identities are often granted broad access to keep deployment paths from breaking, and that access then hardens into an unreviewed default. Oak's article shows that the real question is whether the workload needs persistent reach or whether the privilege only exists because nobody reversed it. Practitioners should treat every permanent grant as a candidate for time-bound replacement.
Usage-based certification is the only review model that reflects reality. Reviewers cannot validate access they never see used, so campaigns that rely on judgment alone create false confidence. The better pattern is to compare granted permissions with observed behaviour and remove rights that never appear in the workload's activity trail. That is how access reviews become evidence, not ceremony.
Source-system retirement is the named concept that separates governance from paperwork. Closing the governance record does not end access if the account still authenticates in the application, cloud service or host. That is the control gap this lifecycle model exposes: a record can say retired while the credential still works. Practitioners need to treat the target system as the source of truth for whether an identity is actually dead.
From our research library:
- NHIs outnumber human identities by 25x to 50x in modern enterprises, according to the Ultimate Guide to NHIs.
- The 2025 Gartner Machine Identity Management in a Hybrid, Automated AI World Survey showed that 32% of organizations use mostly automated methods to manage credentials and only 1% use fully automated methods.
- Read next: NHI Ownership and Accountability Guide
What this signals
Lifecycle enforcement will become the deciding test for machine identity governance programmes. The market is moving away from counts and toward proof that identities are owned, approved, used and retired in the target system. Teams that still separate discovery, rotation and offboarding into unrelated workflows will keep finding credentials that outlive the governance record.
Source-system retirement is the control most programmes still underbuild. Closing an account in the governance tool is not the same as ending authentication in the workload, so the next governance gap is usually the last one to be checked. That pushes practitioners toward controls that verify the identity is dead where it actually operates.
Rotation policy will matter less if governance cannot prove purpose and usage first. Rotation is still necessary, but it does not rescue an identity that has no owner, no purpose and no retirement trigger. The stronger programme design is to treat rotation as one stage in a broader lifecycle rather than as a standalone hygiene task.
For practitioners
- Define a named owner for every machine identity Assign accountability at creation and keep the owner distinct from the technical custodian who manages keys or certificates.
- Block credential issuance until scope is approved Require the approved purpose, system and access scope to exist before any secret, token or certificate is issued.
- Shift machine identity reviews to usage evidence Compare granted access with actual runtime use, then remove unused rights instead of rubber-stamping the certification.
- Rotate credentials by risk, not by calendar alone Shorten rotation intervals for identities with broad or sensitive reach and track each rotation back to the accountable owner.
- Retire identities in the target system first Disable or delete the account where it authenticates, then confirm the credential no longer works before closing the record.
Key takeaways
- Machine identity risk rises when governance treats identities as a list instead of a lifecycle with ownership, approval, usage review and retirement.
- The strongest evidence in the article is the scale: 109 machine identities per human and 40% with access to organizational data.
- A defensible programme ends access in the target system, not just in the governance record, because closed records do not stop live credentials.
Related NHI Mgmt Group resources: Ultimate Guide to NHIs · NHI Ownership and Accountability Guide · Guide to NHI Rotation Challenges · Dropbox Sign breach 2024
Standards & Framework Alignment
This section maps relevant standards and security frameworks to the operational risks and controls described in this guidance.
OWASP Non-Human Identity Top 10 addresses the attack and risk surface, while NIST CSF 2.0, CIS Controls v8 and NIST SP 800-53 Rev 5 set the governance and control requirements practitioners need to meet.
| Framework | Control / Reference | Relevance |
|---|---|---|
| OWASP Non-Human Identity Top 10 | NHI-01 — Improper Offboarding | The article centres on retiring machine identities in the source system, not just closing records. |
| NHI-05 — Overprivileged NHI | Standing admin access and broad reach are a recurring failure mode in the lifecycle model. | |
| NHI-07 — Long-Lived Secrets | Rotation by risk and short-lived credentials directly address the stale-secret problem described here. | |
| Recommendation — Verify that every machine identity is disabled or deleted where it authenticates before closing the record. Reduce persistent privileges and replace broad grants with workload-scoped access. Track secret age and shorten credential lifetimes for machine identities with sensitive access. | ||
| NIST CSF 2.0 | PR.AA-05 — Access Permissions, Entitlements and Authorizations | The lifecycle controls all revolve around approving, reviewing and revoking entitlements correctly. |
| Recommendation — Review and revoke machine identity entitlements based on current authorisation, not historical assignment. | ||
| CIS Controls v8 | CIS-5 — Account Management | The post is fundamentally about assigning, maintaining and removing machine accounts across their lifecycle. |
| Recommendation — Maintain owner records, track account lifecycle events and remove unused machine accounts promptly. | ||
| NIST SP 800-53 Rev 5 | IA-5 — Authenticator Management | Credential issuance, rotation and revocation are central to the governance model described. |
| Recommendation — Enforce credential lifecycle rules for machine identities, including rotation and revocation at source. | ||
Key terms
- Machine Identity: The digital identity of a machine, device, or workload, such as a server, container, or VM, used to authenticate it within a network. Sometimes used interchangeably with NHI, though NHI is the broader category.
- Usage-Based Certification: Usage-Based Certification is an access review method that validates a machine identity’s permissions against how it is actually used. It compares observed runtime activity, such as authentication events and API calls, with assigned entitlements so unused or excessive access can be removed and the account’s scope can be re-certified.
- 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.
- Just-in-Time (JIT) Access: Just-in-Time (JIT) Access grants permissions only when needed and removes them after use. In IAM and PAM, it reduces standing privilege by issuing time-bound access based on approval, context, or policy, limiting exposure for humans, service accounts, and non-human identities while improving auditability and control.
What's in the full article
Oak's full blog post covers the operational detail this post intentionally leaves for the source:
- Stage-by-stage governance examples for creation, approval, rotation, review and retirement
- The evidence model Oak uses to prove machine identity controls are working in the target system
- How Oak handles orphaned identities, dormant access and offboarding lag across legacy systems
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 building or maturing an IAM programme, it is worth exploring.
Published by the NHIMG editorial team on November 16, 2026.
NHI Mgmt Group, the independent authority on Non-Human Identity, IAM, and Agentic AI security. nhimg.org