Skip to content
Human Intelligence seal
Human Intelligence
§ I · Domain MAS · Control 4 of 9
Control
MAS-04
Edition
v.1.3
Status
Public review

MAS · Multi-Agent Systems

MAS-04Privilege escalation prevention

No agent in a chain exercises authority exceeding that of the originating user or the originating agent. Privilege escalation through delegation is prevented at the runtime layer.

When agent A invokes agent B, the invocation should not result in agent B acting with greater authority than agent A had, and ideally not with greater authority than the originating user. MAS-04 forbids the common pattern where a downstream agent operates with broader permissions than the upstream context warrants. Privilege escalation is prevented at the runtime layer and tested as part of red-team coverage under SPC-03.

Table MAS-04.1 · Maturity criteria · v.1.3

L1 to L5 · the lowest level in scope caps the agent
L1Ad hoc

Controls exist informally; documentation is incomplete; no consistent review.

L2Documented

Controls are written down; ownership is assigned; review cadence is defined.

L3Operated

Controls are followed in practice; evidence is collected; deviations are tracked.

Privilege escalation prevention is enforced at the runtime layer; the control is tested as part of adversarial assessment; no privilege-escalation incident has occurred in the review period.

L4Measured

Control effectiveness is measured against documented thresholds; trends are reported.

L5Continuously Improved

Controls evolve based on incident learning; maturity gains are evidenced over time.

The v1.3 edition publishes a control-specific criterion at L3 (Operated). The other levels apply the general ladder to this control; see the maturity ladder.

Evidence an assessor asks for

  1. E.1

    Privilege escalation prevention is enforced at the runtime layer.

  2. E.2

    The control is tested as part of adversarial assessment.

  3. E.3

    No privilege-escalation incident has occurred in the review period.

Table MAS-04.2 · Framework mappings

6 of 7 frameworks carry a control-level reference
NIST AI RMFMANAGE 4.3NIST AI Risk Management Framework 1.0
ISO/IEC 42001A.8.6ISO/IEC 42001:2023 · Artificial intelligence management system
OWASP LLM Top 10·OWASP Top 10 for Large Language Model Applications
SOC 2 (TSC)CC7.4AICPA Trust Services Criteria for SOC 2 Type II
EU AI ActArt. 14(4), Art. 25(1)EU AI Act · Regulation (EU) 2024/1689
MITRE ATLASExecution, Lateral MovementMITRE ATLAS · Adversarial Threat Landscape for AI Systems
SingaporeMAIGF AccountabilitySingapore Model AI Governance Framework / AI Verify / PDPA

Mappings are directional: meeting this control evidences but does not by itself satisfy the external requirement. A middle dot means no control-level reference is published at this edition; see the crosswalk.

Related controls and terms

Cite this control

Human Intelligence. HI-AAF v1.3, MAS-04 Privilege escalation prevention. Public review edition, July 2026. https://humanintel.net/standard/mas/mas-04

Edition v.1.3 · July 2026 · Public review · Markdown · PDF

https://humanintel.net/standard/mas/mas-04

Notice

HI-AAF v1.3 is an independent framework published by Human Intelligence. It is not affiliated with NIST, ISO, AICPA, or any government or accreditation body. The framework is offered openly for review during the current draft cycle.