Saviynt’s identity platform underscores growing NHI governance scope

At a glance

What this is: Saviynt describes an identity platform that spans human and non-human access, with coverage extending into JIT access, MCP Server, and ISPM for AI agents.

Why it matters: That matters because IAM, IGA, PAM, and NHI teams are moving toward shared governance boundaries, and separate control models become harder to defend as identities multiply.

Context

Saviynt is positioning identity governance around a broader access surface than workforce users alone. The article says the platform manages and governs human and non-human access to applications, data, and business processes, while also naming JIT access, MCP Server, and ISPM for AI agents.

For IAM practitioners, the governance question is not whether NHI and agentic access exist, but whether they are being controlled through the same lifecycle, policy, and privilege model as human access. That is the practical implication of an identity platform that spans multiple actor types.

The underlying problem is fragmentation. When workforce IAM, machine identity controls, and agent oversight are handled as separate programmes, accountability and policy consistency become harder to maintain.

Key questions

Q: How should security teams govern human and non-human access in the same programme?

A: They should use one governance model for ownership, approval, review, and revocation, but apply it differently by actor type. Human identities rely on joiner-mover-leaver processes, while NHIs need secrets, certificates, and token lifecycle controls. AI agents add runtime behaviour that must also be monitored. The goal is consistent oversight, not identical workflows for every identity type.

Q: Why do just-in-time access controls matter for non-human identities?

A: JIT matters because it reduces the time a secret, token, or privileged session remains usable. For NHIs, that shortens the blast radius of compromise and reduces unnecessary standing access, but only if the credential lifetime and downstream revocation are enforced as part of the same control.

Q: What breaks when AI agents are reviewed like human users?

A: Human review assumes access is stable long enough to be observed, approved, and recertified. Agentic workflows often complete within one session and can change scope mid-execution, so the review cycle arrives too late to matter. The result is a governance gap where the action has already happened before anyone can certify it.

Q: How should teams decide whether MCP access is safe enough to allow?

A: Teams should allow MCP access only when the agent or server can be bounded with explicit scopes, revocable credentials, and traceable client registration. If the integration depends on static secrets, shared keys, or opaque delegation, the access model is too durable for reliable governance and should be redesigned before production use.

Technical breakdown

Why unified identity governance changes the control boundary

Identity governance is no longer confined to employee accounts and privileged users. When a platform claims coverage across human and non-human access, the technical issue is control boundary convergence: lifecycle, entitlement policy, review, and privilege enforcement have to operate across different identity subjects without creating parallel governance islands. That becomes more complex when AI agents are added, because the access subject may initiate actions dynamically while still depending on identity controls designed for static workloads or people. The practical effect is that access management, governance, and privileged access controls increasingly need a shared policy layer rather than separate administrative stacks.

Practical implication: Treat human IAM, NHI governance, and agent oversight as one access-control programme with different identity subjects, not as disconnected tools.

How JIT access fits machine and agentic identities

Just-in-time access reduces standing privilege by issuing access only when needed, but the model changes when the subject is a service, workload, or AI agent. A human request can be reviewed against a known task, while machine and agentic access may be triggered by system state, workload timing, or runtime orchestration. That means the control value is not only shorter credential duration but also clearer policy on issuance, scope, and revocation across identity types. In practice, JIT becomes a governance pattern only when the privilege granted to a non-human identity is traceable to a specific use case and expires cleanly.

Practical implication: Use JIT policy to constrain non-human and agentic access by task, scope, and expiry, then verify that revocation is automatic and complete.

Why MCP Server coverage matters for agent governance

MCP, or Model Context Protocol, is relevant because it links agents to tools and data sources at runtime. That creates a governance problem if the agent can reach operational systems through identity bindings that were not designed for independent execution paths. The platform naming MCP Server alongside ISPM for AI agents suggests the access layer is being extended toward tool-mediated agent activity, where authorization, auditability, and least privilege must hold even as context changes during execution. The technical challenge is not just authentication, but controlling what an agent can do once it is connected to tools and data sources.

Practical implication: Map every agent-tool connection to explicit authorization scope, and review whether the identity layer can constrain runtime tool use.

Threat narrative

Attacker objective: The objective is to reach sensitive applications, data, or business processes through identity paths that are broader than intended.

1. Entry occurs through legitimate identity connections rather than exploit-driven intrusion, because the control plane is built to govern both people and non-human access.
2. Escalation happens when non-human or agentic identities inherit access patterns that were designed for human workflows and therefore carry broader scope than needed.
3. Impact emerges when fragmented governance cannot consistently answer who or what accessed data, tools, and processes across the full identity estate.

Breaches seen in the wild

• Scania insurance portal breach 2025: An attacker used an external user login, likely stolen by infostealer malware, to take insurance claim documents from a Scania portal.

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

Governance scope is expanding from accounts to actors. The article reflects a broader market reality: identity platforms are being judged on whether they can govern people, service identities, and AI-driven actors in one policy model. That changes the centre of gravity for IAM teams, because identity is no longer just an authentication problem but a lifecycle and authority problem across multiple actor types. Practitioners should now evaluate whether their governance model can follow the actor, not just the account.

Non-human identity controls are becoming part of mainstream identity governance. When a vendor places non-human access, JIT, and AI-agent control in the same narrative as human access, it signals that NHI governance is no longer a niche add-on. The practical implication is that access reviews, entitlement design, and offboarding logic must be able to distinguish workload access from workforce access without relying on separate exceptions. Teams that keep NHI in a side process will continue to accumulate unmanaged privilege.

Agentic access breaks the assumption that entitlement is static at review time. Access review processes were designed for access that persists long enough to be recertified. That assumption weakens when an AI agent can request, use, and release access within a task boundary, because the review window may never see the full privilege state. The implication is that governance increasingly has to move toward issuance-time control and runtime visibility, not only periodic certification.

Runtime governance gap: the article points to a control gap where governance systems are asked to manage runtime access patterns that were not originally designed into human-first IAM models. The issue is not simply more identities, but different identity behaviour, different timing, and different delegation paths. That makes policy consistency, auditability, and revocation discipline the decisive programme questions for security leaders.

Identity convergence is now a platform design test. The market is moving toward platforms that can coordinate IGA, PAM, NHI, and AI-agent governance in one operating model. That does not eliminate specialist control needs, but it does raise the bar for how programmes define ownership, exceptions, and assurance across identity classes. Practitioners should expect governance conversations to shift from tool silos to end-to-end control architecture.

What this signals

Governance programmes will increasingly be measured by actor coverage, not tool count. The practical test is whether one policy model can distinguish users, workload identities, and agents without forcing exceptions into separate silos. That is where identity governance becomes an architecture decision rather than an admin function.

Identity platforms are being pulled toward runtime control. When AI agents and MCP-style tool links enter the same conversation as human access, periodic certification is no longer enough on its own. Security teams need clearer issuance rules, audit trails, and revocation logic for access that can appear and disappear inside one execution path.

For practitioners

• Define one governance model for all identity subjects Map human users, service accounts, workloads, and AI agents into a single entitlement and lifecycle model so policy decisions are consistent across actor types.
• Separate runtime access from review-time access Identify where access is granted and used within one task or session, then move control emphasis to issuance, scope, and revocation rather than periodic recertification alone.
• Inventory MCP-linked tool access Document which agents or automated systems can reach tools and data through MCP-connected paths, and assign explicit authorization boundaries to each path.
• Align JIT policy to non-human identity use cases Require task-scoped justification, expiration rules, and revocation checks for non-human and agentic access instead of reusing human access approval patterns.
• Review offboarding for non-human actors Check whether service accounts, tokens, and agent-linked privileges are removed when their business purpose ends, not only when a person leaves the organisation.

Key takeaways

• The article shows identity governance moving beyond workforce users to include non-human access and agent-oriented control surfaces.
• That expansion matters because access policy, lifecycle control, and privilege oversight now have to work across different actor types in one programme.
• The most important practitioner response is to unify governance while preserving actor-specific rules for issuance, scope, review, and revocation.

Key terms

• Non-Human Identity (NHI): A digital identity assigned to a non-human entity such as a software application, service account, API key, bot, machine, or AI agent that enables it to authenticate and interact with systems without direct human involvement. NHIs now outnumber human identities in most enterprises by 25 to 50 times.
• Just-in-Time Access Request: Just-in-Time Access Request is a pattern that grants access only when it is needed and only for the duration required. It reduces standing privilege by making access temporary, policy driven, and task scoped. This approach is especially useful for contractors, sensitive systems, and short-lived operational work.
• MCP Server: An MCP server is a tool endpoint that connects an AI agent to external systems and data sources through Model Context Protocol. Because it extends what the agent can reach, it becomes part of the identity and access surface and must be reviewed like any other privileged connector.
• Identity Security Posture Management: Identity security posture management is the continuous assessment of identity configuration, privilege, and exposure across an environment. It focuses on drift, overprivilege, and control gaps so teams can see where IAM, PAM, and NHI governance are failing before those gaps become incidents.

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.