Identity attack surface visibility is outpacing IAM, PAM and IGA

At a glance

What this is: Zluri argues that the identity attack surface includes governed and ungoverned identities, entitlements, and access paths, with shadow SaaS and AI tools creating visibility gaps that IAM, PAM, and IGA do not fully cover.

Why it matters: IAM and security teams need a discovery-first view of identity risk because unmanaged applications, orphaned access, and OAuth-linked data paths can sit outside standard governance workflows.

Context

Identity attack surface is the full set of identities, entitlements, access rights, and access paths that exist in an environment, including the ones current IAM, PAM, and IGA tools cannot see. In practice, that means the control plane often covers only the known perimeter, while employees adopt SaaS and AI tools independently and outside formal onboarding.

The article’s central governance gap is not that identity tools are ineffective, but that they were built for a slower procurement model in which IT controlled software intake. Once users can create access paths directly through corporate email, OAuth grants, and self-service sign-ups, the visibility model starts to trail the environment it is meant to govern.

Key questions

Q: What breaks when identity tools only govern sanctioned applications?

A: When IAM, PAM, and IGA only cover sanctioned applications, the environment still contains unmanaged identities, entitlements, and data paths that attackers can use first. The failure is not a missing feature in one product. It is an incomplete control boundary that leaves shadow SaaS, AI tools, and orphaned access outside governance.

Q: Why do shadow AI tools create identity risk for IAM programmes?

A: Shadow AI creates identity risk because hidden tools often inherit access through secrets, service accounts, or delegated APIs without review. That breaks ownership, obscures entitlement scope, and makes revocation slow. IAM teams should treat discovery, ownership, and entitlement mapping as one workflow, not three separate tasks.

Q: How should security teams measure identity attack surface risk?

A: Measure identity attack surface risk by the number of high-risk identities found, the number of exposed credentials revoked, the number of dormant accounts removed, and the reduction in attack paths. Coverage alone is not enough. The most useful metrics show whether hidden identities are being discovered, prioritised, and eliminated before they can be used for lateral movement or compromise.

Q: What should organisations do when visibility tools and governance tools disagree?

A: They should assume the visibility tools are exposing scope the governance tools were never asked to manage, then reconcile that gap before trusting analytics or certification results. The right question is not which tool is right, but which access paths exist outside the current identity model.

Technical breakdown

Why governed identity tools miss the real attack surface

IAM, PAM, and IGA each operate within a defined scope. IAM governs authenticated identities that pass through SSO and lifecycle workflows, PAM protects privileged credentials, and IGA reviews the entitlements it can inventory. The structural limitation is discovery: if an application, token, or account was never onboarded, the tool does not have a governance object to manage. That creates a perimeter problem, not a feature gap. The result is a split between provisioned access and actual access, which attackers exploit first because it is least monitored and least reviewed.

Practical implication: treat discovery as a prerequisite for governance, not a reporting layer after the fact.

How shadow SaaS and AI tools expand identity risk

Shadow SaaS creates unmanaged user accounts and access paths; shadow AI adds a second dimension because many tools can also process enterprise data through connected accounts and OAuth grants. That turns a simple unmanaged application into a data egress channel that may persist after the employee stops using it or leaves the organisation. In identity terms, the risk is not only orphaned access, but also invisible data-sharing permissions that sit outside offboarding and review routines. This is why AI tool adoption changes the shape of the attack surface, not just its size.

Practical implication: map AI and SaaS adoption together, because the same sign-up behaviour can create both access and data exposure.

Why entitlement depth matters more than app count

Counting applications tells you almost nothing about exposure if you cannot see the permissions inside them. A user with read-only access and a user with admin access both appear as a single application in many inventories, yet their blast radius is radically different. The same is true for role drift and accumulated access across job changes. Entitlement depth is where least privilege either holds or breaks, and it is also where IGA programmes are often weakest when application ownership sits outside formal IT control.

Practical implication: inventory application-level entitlements, not just application presence, to find excess privilege hidden inside known tools.

Breaches seen in the wild

• BeyondTrust breach 2024: A stolen BeyondTrust Remote Support API key let a China state-sponsored actor reset accounts and reach US Treasury workstations in 2024.
• Stryker Microsoft Intune Wiper Attack: Compromised Microsoft Intune credentials enable wiper attack wiping 200,000 Stryker devices.

Read and download The State of NHI & AI Agent Breach Report 2026, covering 150+ breaches impacting Non-Human Identities including AI Agents.

NHI Mgmt Group analysis

Identity attack surface visibility is now a governance discipline, not a tooling feature. The article is right to separate visibility from the controls that consume visibility. IAM, PAM, and IGA can only govern what they can discover, which means the programme question is no longer whether the controls exist, but whether the organisation has mapped the identities and access paths those controls never ingest. Practitioners should treat discovery as part of identity governance itself.

Shadow SaaS and shadow AI are structurally different from classic shadow IT because they create unmanaged identity plus unmanaged data flow. A self-provisioned SaaS account is already a governance gap, but an AI tool connected through OAuth can become a persistent data channel as well as an access channel. That combination widens the blast radius and makes ordinary offboarding logic incomplete. The implication is that identity scope now has to include connected data permissions, not only login events.

Entitlement depth is the missing measurement in most identity programmes. Many teams can name the number of applications in their estate, yet cannot show who has admin rights, where role drift has accumulated, or which accounts were created outside formal workflows. That is a classic attack surface blind spot because the unit of risk is not the app, but the permission set inside the app. Practitioners need to reframe reporting from coverage to effective privilege.

Attack surface governance: The useful unit of analysis is the identity path that can actually be abused, not the system that nominally owns it. This article surfaces a category-level shift from managing sanctioned access to finding all access that exists. That shift is important because security assurance built on partial inventories creates false confidence. Teams that continue to measure only connected systems will overestimate control and underestimate exposure.

Visibility must come before intelligence when the underlying dataset is incomplete. Behavioural analytics, risk scoring, and remediation workflows all depend on the access graph they can see. If shadow applications and untracked OAuth grants are missing from that graph, the programme can become more precise about the wrong subset of the environment. The practical conclusion is simple: incomplete discovery makes advanced analytics look better than they are.

From our research library:

• Only 36% of health IT leaders say their organisation applies a privileged access strategy consistently across the enterprise, according to Ponemon Institute research.

What this signals

Identity attack surface has become a programme boundary issue. Teams that still treat IAM, PAM, and IGA as the full identity control plane will miss the growing inventory of access created outside formal procurement. The next maturity step is to discover what the tools never saw, then decide which governance process owns each exposed path.

Discovery must extend into AI-connected access paths. Shadow AI is not just another SaaS category because OAuth-connected tools can carry persistent data access even when active use is low. That means offboarding, entitlement review, and access certification all need scope rules that include connected grants, not just directory-managed accounts.

For practitioners

• Map the unmanaged application estate Build a discovery process that finds SaaS and AI tools adopted outside procurement, using browser signals, finance data, and direct API inspection where available.
• Track application-level entitlements Record who has admin, read-only, and delegated access inside each application so privilege drift is visible at the permission layer, not just the app layer.
• Include OAuth grants in offboarding Remove connected account grants when employees leave or change roles, especially for AI tools that maintain access to email, documents, or internal data sources.
• Separate visibility from analytics Use intelligence platforms only after discovery coverage is established, because scoring models built on partial identity data can hide the size of the blind spot.

Key takeaways

• The article frames identity risk as a visibility problem, not a failure of IAM, PAM, or IGA mechanics.
• Shadow SaaS and AI tools create unmanaged accounts, permissions, and data paths that standard identity programmes often do not inventory.
• The practical response is to discover all access first, then govern entitlement depth, connected grants, and orphaned accounts.

Key terms

• Identity Attack Surface: Identity attack surface is the total set of accounts, tokens, login endpoints, trust paths, and supporting systems that can be probed for access. For password spraying, the risk grows with every externally reachable authentication path and every dormant or weakly protected identity.
• Shadow SaaS: Shadow SaaS is the set of unauthorised or unreviewed software-as-a-service tools used outside central security governance. These applications often bypass normal identity controls, making them difficult to inventory, monitor, and harden against credential-based abuse.
• Shadow AI: AI agents, copilots, or connected tools operating without full visibility or governance from security teams. Shadow AI becomes an identity problem when those systems authenticate with unmanaged tokens, service accounts, or OAuth apps that can reach production resources.
• Entitlement Depth: Entitlement depth is the level of detail an identity platform can report and govern beyond simple app access. It includes roles, privileges, delegated permissions and inherited access, which are the elements that determine real blast radius and audit quality.

Deepen your knowledge

NHI governance, identity lifecycle, and secrets management 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.