A governance failure where multiple AI agents share one role or ServiceAccount, causing their permissions to merge into a single blast radius. The problem is structural, not just operational, because the audit trail and accountability model stop matching the real behavioural boundaries.
Expanded Definition
Shared-identity collapse occurs when multiple AI agents, automation jobs, or service workflows operate under one role or ServiceAccount, so their permissions, logs, and failure domains become indistinguishable. In NHI governance, that is not just a convenience issue; it breaks the mapping between NIST Cybersecurity Framework 2.0 identity controls and the actual behaviour of the workload.
Definitions vary across vendors on whether the issue is treated as an identity design flaw, an agentic AI governance failure, or a privileged access problem, but the operational risk is consistent: one shared credential can expand access across unrelated actions. NHI Management Group treats it as a structural NHI control failure because the identity boundary no longer reflects the real agent boundary, especially in systems that mix tool use, delegation, and ambient permissions. This is closely related to the visibility and lifecycle problems discussed in the Ultimate Guide to NHIs. The most common misapplication is assigning a single ServiceAccount to multiple agents for convenience, which occurs when teams optimise deployment speed over per-agent accountability.
Examples and Use Cases
Implementing agent identity rigorously often introduces more provisioning and policy overhead, requiring organisations to weigh clean accountability against faster rollout of automation.
- Two code-review agents share one Kubernetes ServiceAccount, so a prompt-injection event on one agent can inherit the same cluster permissions as the other.
- An internal ticket triage agent and a deployment agent use the same API key, making it impossible to prove which agent triggered a destructive action.
- A batch of copilots runs under one cloud role, and one agent’s tool misuse expands into all connected data sources because the identity is not separated per function.
- A security workflow reuses the same secret across staging and production assistants, which obscures which environment actually initiated a risky call.
- The pattern seen in incidents such as the 52 NHI Breaches Analysis often shows that shared credentials make containment slower and root-cause analysis weaker.
These scenarios are especially dangerous in agentic systems because execution authority and tool access are dynamic, while shared identities assume a static trust boundary. Guidance from NIST Cybersecurity Framework 2.0 supports separating access functions so attribution remains meaningful.
Why It Matters in NHI Security
Shared-Identity Collapse turns a manageable compromise into an enterprise-wide blast radius. Once one agent is over-permissioned, every other agent using the same role inherits that exposure, and incident responders lose the ability to isolate intent, sequence, or ownership. That creates governance gaps in least privilege, segregation of duties, and auditability, which are all central themes in the Top 10 NHI Issues. It also undermines Zero Trust efforts, because trust decisions rely on a clear identity boundary rather than a pooled credential.
NHIMG data shows that 97% of NHIs carry excessive privileges, which means identity reuse is often paired with overbroad access and makes collapse even harder to contain. The same risk pattern appears in breach reporting where secrets are exposed in code, pipelines, and tooling, as described in the Ultimate Guide to NHIs. Practitioners need to treat shared identity as a sign that the operating model has already drifted away from the control model. Organisations typically encounter the consequence only after one agent is compromised or misfires, at which point identity separation becomes operationally unavoidable to address.
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, OWASP Agentic AI Top 10 and CSA MAESTRO address the attack and risk surface, while NIST CSF 2.0 and NIST Zero Trust (SP 800-207) set the governance and control requirements practitioners need to meet.
| Framework | Control / Reference | Relevance |
|---|---|---|
| OWASP Non-Human Identity Top 10 | NHI-01 | Shared identities collapse attribution and expand blast radius across NHIs. |
| OWASP Agentic AI Top 10 | A-03 | Agentic systems need identity separation to preserve tool-use accountability. |
| NIST CSF 2.0 | PR.AC-4 | Access permissions must be managed to keep identity boundaries meaningful. |
| NIST Zero Trust (SP 800-207) | AC-1 | Zero Trust requires explicit identity and least-privilege enforcement per workload. |
| CSA MAESTRO | ID-02 | Agent security guidance emphasizes distinct identities and scoped delegation. |
Separate agent identities and scope delegation so one agent cannot inherit another's privileges.
Related resources from NHI Mgmt Group
- How should regulated teams decide between shared SaaS and tenant-owned identity platforms?
- How should security teams reduce risk from shared secrets in identity systems?
- When should organisations choose full isolation over shared identity services?
- Why do shared patient records create new identity governance risks?
Deepen Your Knowledge
Reviewed and updated by the NHIMG editorial team on August 20, 2026.
NHI Mgmt Group — the #1 independent authority on Non-Human Identity, IAM, and Agentic AI security. nhimg.org
