Join our Newsletter — 33% off our NHI Course
Home› Glossary› Governance, Ownership & Risk› Evidence Half-Life
Governance, Ownership & Risk

Evidence Half-Life

← Back to Glossary
By NHI Mgmt Group Updated October 23, 2026 Domain: Governance, Ownership & Risk

Evidence Half-Life is the period after which security evidence loses enough freshness to become unreliable for decisions. In IAM, NHI governance, and incident response, logs, attestations, scans, and approvals age quickly as systems, identities, and access states change, so older evidence may no longer support current risk or compliance conclusions.

What Evidence Half-Life Means

Evidence Half-Life describes how quickly security evidence stops being trustworthy enough for decision-making. Freshness matters because a log, scan, approval, or attestation can be correct when collected and misleading shortly after if systems, identities, access states, or controls have changed.

The idea is especially important in IAM, NHI governance, and incident response, where evidence is often used to justify access decisions, prove control operation, or reconstruct what happened. The value of the evidence is not just whether it exists, but whether it still reflects the current environment.

Why Freshness Determines Trust

Security teams often treat evidence as if it were a permanent fact. In practice, evidence is a snapshot. A successful review today may already be stale tomorrow if a service account is rotated, an integration is changed, a privilege is added, or a system drifts out of baseline.

This is why evidence half-life is not only about time passed, but about change rate. Fast-moving environments, high-privilege systems, and delegated access paths all shorten the useful life of evidence because the underlying reality can diverge quickly from the record.

Where Evidence Ages Fastest

Evidence decays fastest when it is tied to states that change often: credentials, sessions, tokens, approvals, runtime permissions, configuration drift, and ephemeral infrastructure. A point-in-time export may look authoritative while missing changes that happened immediately after collection.

That creates a common governance trap. Teams may rely on a clean scan, an old approval trail, or an archived log set and assume the conclusion still holds. The older the evidence, the more likely it is to describe a previous security state rather than the current one.

How to Interpret Old Evidence

Old evidence is not useless, but it should be treated as historical context rather than current proof unless the question being answered is also historical. For current assurance, the reader should ask whether the evidence is still within the operational window in which the environment is stable enough for the finding to remain valid.

Evidence half-life is therefore a judgment about confidence. Short half-life evidence supports near-term operational decisions; long-lived evidence is better for trend analysis, retrospective investigations, or control history. The critical distinction is whether the evidence still matches the present control state.

Risk and Threat Considerations

Stale evidence can create false confidence, especially when it is used to approve access, close an incident, or certify compliance. Attackers benefit from that gap because evidence that is merely old can conceal privilege changes, lingering sessions, credential abuse, or recovery gaps.

Failure mechanism: The control or review process relies on evidence that no longer matches the live state, so a material change slips through without revalidation.

Impact: Organisations can misjudge exposure, miss active compromise, retain excessive access, or sign off on controls that are no longer effective.

Standards & Framework Alignment

This section maps relevant standards and security frameworks to the operational risks and controls described in this guidance.

NIST CSF 2.0 and NIST SP 800-53 Rev 5 set the governance and control requirements practitioners need to meet.

FrameworkControl / ReferenceRelevance
NIST CSF 2.0GV.OV-01 — Oversight of cybersecurity risk management activitiesEvidence half-life affects how current oversight evidence remains for risk decisions.
Recommendation — Review evidence freshness before using it to support oversight decisions.
NIST SP 800-53 Rev 5AU-6 — Audit Record Review, Analysis, and ReportingStale audit evidence can undermine analysis of current events and security conditions.
CA-2 — Control AssessmentsAssessments depend on evidence that still reflects the control state being evaluated.
IA-5 — Authenticator ManagementCredential and authenticator state changes rapidly, shortening the useful life of related evidence.
Recommendation — Analyze audit records within a freshness window that matches the decision being made. Use recent evidence when assessing whether controls are operating effectively. Revalidate authenticator-related evidence after credential changes or rotation.

Practitioner Guidance

Why practitioners should care: Evidence half-life should shape how long a team is willing to trust logs, scans, approvals, and attestations for a given decision. The shorter the half-life, the more often evidence must be refreshed before it can support access, assurance, or incident conclusions.

Common misunderstanding: Teams often assume that a documented control outcome remains valid until the next scheduled review. In reality, the evidence may expire operationally long before the review date if the underlying environment changes quickly.

Practitioner takeaway: Treat evidence as time-sensitive proof, not a permanent record, and match its reuse to the volatility of the system and decision it is supposed to support.

Free weekly newsletter

Subscribe to the NHI & AI Identity Journal

The latest on NHI and Agentic AI security – articles, research, breaches, news and events every week.

Bonus 33% off our NHI Course when you subscribe.

NHIMG Editorial Note
Reviewed and updated by the NHIMG editorial team on October 23, 2026.
NHI Mgmt Group — the #1 independent authority on Non-Human Identity, IAM, and Agentic AI security. nhimg.org