Ownership should be shared across IAM, security, and finance, because the problem crosses entitlement control, data exposure, and budget management. IAM can define and revoke access, security can evaluate risk and shadow AI, and finance can monitor consumption and renewal exposure. No single team can close the gap on its own.
Why This Matters for Security Teams
AI token usage and SaaS sprawl are not just procurement issues. They create overlapping risk across identity, data access, and spend, which means the wrong owner often leaves gaps in revocation, monitoring, and renewal control. NIST’s Cybersecurity Framework 2.0 treats governance as an enterprise function, not a siloed technical task, and that framing fits this problem well.
In practice, token sprawl is how shadow integrations, unmanaged OAuth grants, and duplicated SaaS subscriptions accumulate before anyone notices. NHIMG has repeatedly highlighted the operational impact in the Guide to the Secret Sprawl Challenge and the Top 10 NHI Issues, where fragmented ownership is a recurring root cause. The decision surface is wider than access review alone because a token can authorize data movement, automation, billing exposure, and third-party persistence at the same time.
NHIMG research on the State of Secrets in AppSec shows organisations are dedicating an average of 32.4% of security budgets to secrets management and code security, which is a strong signal that this is already an enterprise-scale governance problem. In practice, many security teams encounter the true cost only after renewals, excess tokens, and dormant SaaS accounts have already multiplied.
How It Works in Practice
Ownership should be shared, but accountability should be explicit. IAM typically owns issuance standards, lifecycle rules, and revocation for tokens tied to user or workload identity. Security owns risk policy, detection, and exception handling for shadow AI, excessive scopes, and suspicious SaaS connections. Finance owns spend control, renewal oversight, and vendor consolidation. The operating model works best when one function is named the decision coordinator, even if the control work is distributed.
Current guidance suggests building a single inventory that joins identity data, SaaS contracts, API keys, OAuth grants, and billing records. That inventory becomes the basis for review rather than relying on separate lists from procurement and security. When an AI tool requests access, the question is not only whether the token exists, but whether the service is approved, the scope is minimal, the renewal date is visible, and the business owner can justify continued use.
- IAM defines token issuance policy, scope limits, and revocation triggers.
- Security monitors for anomalous token use, over-permissioned apps, and shadow AI services.
- Finance tracks subscriptions, unused seats, duplicate tools, and renewal risk.
- App owners validate business need and approve exceptions with an expiry date.
This is where lifecycle discipline matters. NHIMG’s Ultimate Guide to NHIs — Lifecycle Processes for Managing NHIs is useful because tokens and SaaS entitlements should be treated as managed identities with birth, use, review, and retirement stages. That aligns with enterprise governance better than ad hoc app-by-app approval. These controls tend to break down when SaaS procurement is decentralized across business units because no single system sees both the risk and the recurring spend.
Common Variations and Edge Cases
Tighter governance often increases process overhead, requiring organisations to balance speed of adoption against control over token issuance and SaaS renewal. That tradeoff is especially visible in fast-moving teams that rely on AI copilots, browser extensions, and self-serve SaaS trials. Best practice is evolving, but current guidance suggests creating tiered approval paths instead of one heavy gate for every tool.
There is also a genuine edge case where finance sees spend but not technical risk, while security sees risk but not contract terms. In those environments, governance fails unless the organisation defines a shared intake and a common risk taxonomy. The 2024 ESG Report: Managing Non-Human Identities is a reminder that governance maturity is uneven, and the organisations with the weakest visibility are often the ones with the most fragmented identity surface.
Where the model breaks down is highly decentralized SaaS buying, especially when teams approve AI tools directly through credit cards or browser plugins. In those cases, finance discovers spend after the fact, security discovers exposure after a token leak, and IAM never receives the entitlement into a managed workflow. The practical fix is not to choose one owner, but to formalize shared accountability with a single system of record and clear escalation rules.
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 address the attack and risk surface, while NIST CSF 2.0 and NIST AI RMF set the governance and control requirements practitioners need to meet.
| Framework | Control / Reference | Relevance |
|---|---|---|
| OWASP Non-Human Identity Top 10 | NHI-03 | Token sprawl is an NHI lifecycle and revocation problem. |
| NIST CSF 2.0 | GV.OC, PR.AC | This is an enterprise governance and access control ownership issue. |
| NIST AI RMF | GOVERN | AI token usage needs accountability, oversight, and risk management. |
Assign cross-functional governance, then enforce least-privilege access and periodic review.
Related resources from NHI Mgmt Group
Deepen Your Knowledge
Reviewed and updated by the NHIMG editorial team on June 24, 2026.
NHI Mgmt Group — the #1 independent authority on Non-Human Identity, IAM, and Agentic AI security. nhimg.org
