Task Delegation
The transfer of responsibility for achieving an outcome from one agent to another, where the receiving agent determines how the outcome is reached.
- StatusIndustry term
- AdoptionEmerging
- LayerGovernance
Why it matters
Delegation between agents is delegation of authority, whether or not the framework models it that way. The moment agents delegate across organizational boundaries — which A2A (Agent2Agent) is explicitly designed to enable — 'who decided' stops being answerable inside any single institution.
Where it comes from
A2A protocol; long-standing in multi-agent systems research.
Defining sources
- Agent-to-Agent (A2A)Originated at Google; contributed to the Linux FoundationOfficial text In the standards register
What it is not
These are routinely confused with Task Delegation. The distinctions have consequences for how a system is governed.
- Tool Invocation
Invocation specifies the how. Delegation specifies only the what and leaves the how to the receiver.
- Escalation
Delegation passes work outward or downward. Escalation passes a decision upward for a judgment the delegate is not permitted to make.
How it connects
What it needs
- Task Delegation is limited by Trust Boundary Canon term
- Task Delegation depends on A2A Standard
- Task Delegation is limited by Agent orchestration Canon term
What relies on it
- Task Delegation delegates authority to Agent Identity Canon term
MARQUE questions it addresses
- AccountabilityWhen it goes wrong, who answers?
Related framework objectives
Connected through the MARQUE (a reference framework organized around eight questions) crosswalk, an editorial mapping. These objectives from the Financial Services AI Risk Management Framework are assigned to the same questions that this term addresses. The ones sharing most of this term's vocabulary are shown first, as a reading aid. The assignment is to the question, not to the term: the source does not map this term to any objective directly.
Question: Accountability · 55 control objectives assigned
- GV-1.5.1AI Performance and Risk MonitoringGovern
- GV-2.3.3AI Governance StructureGovern
- GV-6.1.5Third-Party AI Risk DocumentationGovern
Cite and reuse
Cite the identifier QIS-TERM-00008, not the URL: identifiers are stable, URLs may change. This entry is free to read, quote and index under the dual license. Corrections sent to editor@qisfund.com are published.
Technical details: identifier, graph relationships and record
| Predicate | Target identifier | Target | Predicate meaning |
|---|---|---|---|
delegatesTo | term:agent-identity | Agent Identity | The subject transfers authority to the object. |
constrainedBy | term:trust-boundary | Trust Boundary | Read the other way round: the object sets limits on the subject. The subject bounds what the object may do. |
addresses | marque:accountability | Accountability | The subject speaks to the object as a question or concern. |
dependsOn | std:a2a | A2A | The subject cannot function correctly without the object. |
relatedTo | term:context-exchange | Context Exchange | An association too weak or too general for a stronger verb. |
constrainedBy | term:orchestration | Agent orchestration | Read the other way round: the object sets limits on the subject. The subject bounds what the object may do. |
- Canonical identifier
QIS-TERM-00008- Tier
- Canonical industry term
- Domain · layer
- Capability · Governance
- Semantic aliases
- None recorded
- First published
- 2026-08-02
- Last reviewed
- 2026-08-02 · 180-day cycle