Delegated identity is when one actor acts on behalf of another with explicit permission and bounded authority. In AI-assisted commerce, it requires clear consent, limited scope, and traceable records so the retailer can distinguish authorised delegation from unauthorised automation.
Expanded Definition
Delegated identity is the security pattern that lets one identity act for another with explicit consent, bounded authority, and auditable traceability. In NHI operations, it shows up when an AI agent, service account, or application is allowed to initiate actions on behalf of a user or system under tightly defined rules.
Definitions vary across vendors, especially in commerce, customer support, and agentic workflows, where “delegation” can mean anything from proxy login to policy-scoped transaction execution. NHI Management Group treats delegated identity as narrower than generic automation: the delegating party must remain identifiable, the delegated actor must have limited rights, and the resulting activity must be attributable. That distinction matters because NIST Cybersecurity Framework 2.0 emphasizes governed access and traceability, not just functional convenience.
The most common misapplication is treating delegated identity as a standing shared account, which occurs when organisations grant broad, long-lived permissions without proving who authorised the action or when that authority expires.
Examples and Use Cases
Implementing delegated identity rigorously often introduces workflow friction, requiring organisations to weigh user convenience against tighter approval, logging, and revocation controls.
- An AI shopping assistant places an order on behalf of a customer, but only within a pre-approved budget and product category, with every transaction tied back to the original consent.
- A helpdesk agent uses delegated access to reset a locked account, but the privilege is time-bound and approved through Ultimate Guide to NHIs style lifecycle controls rather than permanent elevation.
- A developer tool requests repository changes through a service identity that is authorised only for specific branches, similar to the failure patterns described in the JetBrains GitHub plugin token exposure analysis.
- An AI procurement agent submits a renewal request, but policy requires step-up approval before payment execution and retains a record of the human principal behind the delegation.
- A federated workload receives temporary authority to call another internal API, using short-lived credentials and policy checks aligned to NIST Cybersecurity Framework 2.0.
Why It Matters in NHI Security
Delegated identity is where accountability either survives automation or disappears into it. When the delegated actor is an NHI or AI agent, the organisation must prove not only that the action was technically possible, but that it was intentionally authorised, constrained, and revocable. Without that discipline, delegated access becomes an attack path for privilege creep, unauthorised tool use, and hidden lateral movement.
This is especially relevant because NHI risk is often underestimated: according to the Ultimate Guide to NHIs, 97% of NHIs carry excessive privileges, increasing unauthorised access and broadening the attack surface. Delegated identity can reduce that exposure only when paired with role-based boundaries, just-in-time elevation, and revocation discipline. It also supports lessons reflected in the 52 NHI Breaches Analysis, where weak control of machine-held authority repeatedly magnified impact.
Organisations typically encounter delegated identity as an urgent issue only after an agent, token, or proxy path is used in an incident, at which point attribution and scope control become 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 and OWASP Agentic AI Top 10 address the attack and risk surface, while 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 | Delegated identity depends on scoped machine access and traceable authorization. |
| OWASP Agentic AI Top 10 | AGENT-03 | Agentic workflows must constrain what an AI agent can do on another party's behalf. |
| NIST Zero Trust (SP 800-207) | Policy Decision Point | Zero Trust requires continuous policy checks before delegated access is allowed. |
Use time-bound, least-privilege delegated access with clear owner attribution and revocation.
