Harm Context Standard - (HCS)
OOF® Origin Open Foundation®
Independent Methodological Authority
Architecture Ecosystem: Structured Reality Standards™
Architecture Family: HAGA™ — Harm Governance Architecture
Operational Layer: Harm Context
Category: Governance & Enforcement
Subcategory: Harm Governance
Type: Parent Standard
Governed Space: Harm Context
Standard ID: HAGA-HCS-001
Origin ID: OOF.PS.16.01.00.01
Version: 1.0
Status: Canonical · Open Standard
Effective Date: 12 September 2026
Compatibility: OOF® Methodology OS™ · Modular OOF® Architecture Ecosystem
AI-Readable: Yes
Authority: OOF®
Protection: MIP® — Methodological Intellectual Property
Canonical Language: English (UCL™)
Canonical Definition
The Harm Context Standard (HCS) governs how the context within which Harmmay exist, emerge, persist, change, or become relevant for governance is
identified, bounded, structured, maintained, and made sufficiently
explicit before downstream Harm governance proceeds.
HCS establishes the governed relationship between a potentially affected
subject or object, the environment in which Harm is relevant, the
applicable Harm domain, the temporal conditions, material relationships,
and other context necessary to understand what Harm governance is actually
about.
The Standard governs the transition:
Harm-Relevant Reality → Harm Context Identification → Affected Subject /
Object Context → Domain & Temporal Context → Context Relationships →
Governed Harm Context
HCS governs the context of Harm.
It does not establish that Harm exists.
Operational Role
Harm cannot be governed reliably when the context in which it is alleged,suspected, emerging, or observed remains undefined.
The same operational occurrence may represent radically different Harm
relevance depending upon who or what is affected, where the occurrence
takes place, over what period, under what conditions, within which domain,
and in relation to which surrounding circumstances.
A physical contact between a human and a machine may be ordinary
interaction, a near miss, an exposure condition, or part of actual Harm.
An AI-generated output may be harmless in one context and materially
harmful in another.
A financial loss may represent Harm to one subject while being an ordinary
market consequence for another.
A condition that produces no immediately observable effect may
nevertheless become relevant to medical, environmental, cognitive,
systemic, or accumulated Harm.
HCS therefore prevents downstream HAGA™ governance from beginning with an
undefined assumption that an occurrence, action, condition, output,
failure, interaction, or allegation already constitutes Harm.
Its core responsibility is:
to establish the governed context necessary to determine what Harm
governance concerns before Harm itself is treated as established.
Core Question
Within what context can Harm meaningfully exist and become governable?Governance Flow
Harm-Relevant Reality → Context Trigger → Context Identification →Affected Subject / Object Context → Harm Domain Context → Temporal Context
→ Material Context Relationships → Context Sufficiency → Governed Harm
Context
Where context changes materially:
Governed Harm Context → Material Context Change → Context Reassessment →
Updated Governed Harm Context
Where context is insufficient:
Harm-Relevant Signal → Insufficient Context → Context Qualification /
Preservation → Pending Harm Context
Governed Object
The governed object of HCS is Harm Context.Harm Context is the structured set of circumstances necessary to
understand the environment within which Harm may become relevant as a
distinct governed object.
Harm Context may concern:
- an individual
- multiple individuals
- a defined group
- an organization
- a physical object
- a digital object
- an operational system
- infrastructure
- an environment
- a population or society
- combinations or relationships among them
The existence of a Harm Context does not establish the existence of Harm.
Governance Trigger
HCS may become applicable when an occurrence, condition, interaction,state, signal, allegation, observation, exposure possibility, or other
governed information creates a legitimate reason to establish
Harm-specific context.
A trigger may originate from human activity, organizational processes,
physical systems, digital systems, AI systems, autonomous systems,
environmental conditions, multi-system interactions, or other operational
mechanisms.
The trigger need not itself be harmful.
HCS therefore recognizes:
Harm Governance Trigger ≠ Harm Emergence
and:
Harm-Relevant Occurrence ≠ Established Harm
A near miss, unsafe condition, anomalous AI output, physical contact,
unusual system behavior, adverse signal, complaint, or suspected exposure
may justify Harm Context governance without justifying a Harm
Determination.
Harm Context Formation
Harm Context must be formed around the Harm-relevant question rather thanaround every fact available about the surrounding Operational Reality.
Context formation should identify the circumstances materially necessary
to understand:
- what may be affected
- what kind of Harm may be relevant
- when the relevant circumstances exist or existed
- where relevant boundaries may lie
- which relationships materially affect Harm interpretation
- whether the context concerns an individual, collective, systemic,
or
environmental condition; - whether Harm may be discrete, continuous,
accumulated, latent, or propagated; - what remains unknown.
HCS does not require complete knowledge of every surrounding circumstance.
It requires sufficient Harm-specific context for the next legitimate
governance step.
This creates the governing principle:
Complete Operational Knowledge ≠ Required Harm Context
and equally:
Available Information ≠ Sufficient Harm Context
Affected Subject & Object Context
HCS must identify who or what may be affected without prematurelydetermining that the identified subject or object has actually been
harmed.
The affected context may involve a single identifiable subject, multiple
subjects, a defined group, a system, infrastructure, an environmental
object, or a broader population.
HCS must also allow the affected population to evolve.
A Harm-relevant condition initially associated with one subject may later
reveal additional affected subjects.
A systemic Harm may become recognizable without one single subject
representing the complete Harm.
A collective Harm may arise from repeated effects distributed across many
individuals.
An environmental Harm may exist without a conventional human claimant.
Therefore:
Single Affected Subject ≠ Required Harm Structure
and:
Absence of a Single Harmed Individual ≠ Absence of Governable Harm Context
Subject identification must remain distinct from identity governance.
Where origin-bound identity, ownership, authority, permission, provenance,
or identity continuity is material, HCS may consume appropriately governed
identity information without assuming the functions of OBIDENITY®.
Harm Domain Context
HCS must establish the domain or combination of domains within which Harmmay become meaningful.
Relevant Harm domains may include physical, medical or health, digital,
financial, property, environmental, organizational, cognitive, societal,
or other sufficiently defined domains.
Domain identification must not become a rigid assumption that one Harm
belongs permanently to one domain.
A physical Harm may produce medical, financial, organizational, or
psychological consequences.
A digital Harm may propagate into financial or operational Harm.
A cognitive Harm may produce physical consequences.
An environmental Harm may create distributed human and economic
consequences.
HCS therefore supports:
Single-Domain Context
Multi-Domain Context
Cross-Domain Context
and, where appropriate:Domain Transition
The purpose is not to classify every consequence in advance.The purpose is to ensure that Harm governance does not silently apply the
assumptions of the wrong domain. Temporal Context & Harm Lifecycle
Position
Harm Context must preserve relevant time.
The relevant temporal structure may include:
- beginning of Harm-relevant conditions
- duration of exposure-relevant circumstances
- time of an identifiable occurrence
- periods of continuous activity
- accumulation periods
- delayed recognition
- latency
- persistence
- recurrence
- historical conditions
- material context changes
The moment when Harm becomes visible is not necessarily the moment when
Harm began.
Therefore:
Recognition Time ≠ Harm Emergence Time
and:
Observation Time ≠ Harm Origin Time
HCS must preserve sufficient temporal context to allow HAGA™ downstream
standards to distinguish discrete, continuous, accumulated, latent, and
propagated Harm.
It must not fabricate an exact temporal boundary where evidence supports
only a range or uncertain period.
Temporal Precision Must Not Exceed Temporal Evidence Operational Mechanism
Context
Harm may arise through human action, organizational process, mechanical
operation, software execution, AI behavior, autonomous operation, human-AI
interaction, multi-agent activity, environmental conditions, or
combinations of mechanisms.
HCS may preserve which operational mechanism is relevant to Harm Context.
It does not govern that mechanism.
For example:
AI Output → Harm-Relevant Context
does not make HCS an AI behavior standard.
Autonomous Robot Action → Harm-Relevant Context
does not make HCS an autonomous execution standard.
Industrial Failure → Harm-Relevant Context
does not make HCS an Operational Reality architecture.
The governing distinction is:
Operational Mechanism Context ≠ Operational Mechanism Governance
HCS consumes only the mechanism information materially necessary to
establish Harm Context. Perception-to-Action Context
Embodied and autonomous systems may create Harm-relevant circumstances
without passing through linguistic reasoning.
An operational chain may take the form:
Perceive → Interpret → Select → Authorize → Act → Observe → Correct
HCS must therefore remain capable of establishing Harm Context even when
there is no prompt, text output, linguistic decision, or human-readable
reasoning event corresponding to the relevant action.
The Harm Context may instead depend upon operational states such as
perception, movement, proximity, authorization, physical interaction,
environmental response, or resulting subject state.
HCS does not determine whether those autonomous states were correctly
produced.
It establishes only their relevance to Harm Context.
Absence of Linguistic Reasoning ≠ Absence of Harm-Relevant Context
Physical Interaction & Touch Context
Physical AI, robotics, autonomous vehicles, industrial systems, medical
robotics, cobots, and other embodied systems may create Harm-relevant
context through direct or indirect physical interaction.
Relevant context may include:
Approach → Proximity → Contact → Force / Pressure / Motion → Physical
Effect → Subject / Environmental Response
HCS may preserve contact state, force state, pressure state, movement,
duration, affected subject state, environmental conditions, or similar
information where materially relevant.
But physical interaction alone does not establish Harm.
Physical Contact ≠ Harm
Force ≠ Harm
Machine Action ≠ Harm
The operational mechanism may be governed through ASGA™, ORA™, or another
applicable architecture.
HCS establishes only the Harm-specific context created by that
interaction. Continuous, Accumulated & Latent Context
HCS must not assume that Harm requires a single incident.
Some Harm contexts develop through continuing conditions.
Others arise from repeated low-level effects.
Others remain materially invisible until later consequences emerge.
HCS must therefore support context structures for:
Discrete Harm-Relevant Conditions
Continuous Harm-Relevant Conditions
Accumulated Harm-Relevant Conditions
Latent Harm-Relevant Conditions
Propagated Harm-Relevant Conditions
A context may transition between these forms.A repeated condition initially treated as isolated may later reveal
accumulation.
A localized effect may later reveal propagation.
A condition believed resolved may later reveal latency.
HCS must preserve enough context for these changes to remain governable.
Context Relationships & Dependency Boundary
Harm Context may depend upon relationships between subjects, systems,
organizations, environments, processes, actions, outputs, conditions, and
historical states.
HCS governs only those relationships insofar as they are necessary to
establish Harm Context.
It does not convert those relationships into findings of causation,
responsibility, accountability, ownership, or liability.
A relationship may be:
Harm-Relevant
without being:Causally Established
and a causal relationship may be relevant without establishing:Accountability or Liability
Therefore:
Context Relationship ≠ Causal Determination
Causal Relevance ≠ Accountability
Accountability ≠ Liability
This boundary allows HAGA™ to receive information from ORA™, AGA™, ADIT®,
ASGA™, AIG®, OBIDENITY®, or other appropriate governance systems without
absorbing their governed domains. Context Uncertainty & Competing Contexts
HCS must support uncertainty rather than forcing premature certainty.
At the beginning of Harm governance, relevant context may be incomplete,
disputed, contradictory, evolving, or supported by multiple plausible
interpretations.
Possible context conditions may include:
Sufficiently Established
Partially Established
Conditionally Established
Disputed
Incomplete
Pending
Unable to Determine
Where multiple plausible contexts remain materially possible, HCS shouldpreserve them until sufficient governance supports consolidation or
exclusion.
One Available Narrative ≠ Governed Harm Context
HCS must not select a convenient context merely because it simplifies
downstream determination.
Context Sufficiency
Harm Context becomes sufficient when it provides enough governed structurefor the next legitimate HAGA™ lifecycle step without requiring unsupported
assumptions.
Context Sufficiency is purpose-dependent.
The context sufficient to identify Harm Exposure may be less extensive
than the context required for Harm Determination.
The context sufficient to contain a potentially dangerous condition may
also differ from the context required to establish final Harm boundaries.
Therefore:
Context Sufficiency ≠ Context Completeness
and:
Sufficient for Exposure Governance ≠ Sufficient for Harm Determination
HCS must preserve what the context currently supports and what remains
unresolved.
Context Change & Reassessment
Harm Context is not permanently fixed at first establishment.Material changes may include:
- newly identified affected subjects
- newly discovered operational conditions
- newly recognized domains
- changed temporal understanding
- newly available evidence
- newly identified propagation
- previously latent consequences
- corrected Operational Reality
- changed relationship information
- previously unknown historical conditions
Where such changes materially affect Harm governance, the context must be
reassessed.
The governing flow is:
Governed Harm Context → Material New Information → Context Impact
Assessment → Context Reassessment → Preserved Prior Context + Updated
Context State
A new context state must not silently erase the earlier governed state.
Historical Context Integrity
HCS must preserve material historical context where later information
changes how Harm is understood.
If the original context was:
Single Subject + Discrete Event
and later evidence establishes:
Multiple Subjects + Accumulated Condition
the earlier context should remain reconstructable.
This protects HAGA™ from retrospective rewriting.
Current Harm Context ≠ Historical Harm Context
Historical context may be essential for understanding recognition delay,
exposure duration, accumulation, propagation, containment decisions,
mitigation actions, and eventual resolution. Cross-Architecture Boundary
HCS operates at the Harm-specific context layer.
ORA™ may establish relevant Operational Reality.
ASGA™ may establish relevant autonomous operational states.
AIG® may govern relevant AI behavior.
CLIA® may govern relevant cognition.
MGIA™ may govern relevant memory.
OBIDENITY® may govern relevant origin-bound identity relationships.
ADIT® may govern supporting evidence and auditability.
INTEGROS® may govern relevant integrity states or deviations.
VALIDOS™ may govern formal validation where required.
AGA™ may govern accountability relationships.
HCS may consume relevant outputs from these architectures.
It does not recreate them.
The canonical handoff principle is:
Governed External Context → Harm-Specific Contextualization → HCS
not:
HCS → Reconstruct Every Upstream Governance Domain Lifecycle Entry
Condition
HCS begins when there is sufficient reason to treat Harm Context as a
distinct governance requirement.
This may occur before Harm exists, before Harm is recognized, or before
sufficient information exists to determine whether Harm occurred.
HCS therefore may legitimately begin with:
Potential Harm Relevance
without producing:Established Harm Lifecycle Exit & Handoff
HCS reaches its principal handoff when Harm Context is sufficiently
governed to support the next legitimate HAGA™ lifecycle stage.
The normal architecture flow is:
Governed Harm Context → HES — Harm Exposure Standard
However, Harm governance does not require artificial sequential delay
where Harm is already emerging or recognized.
Where Operational Reality already indicates potential existing Harm, the
governed context must remain available to downstream HAGA™ standards
without forcing the system to pretend that the lifecycle is only beginning
at exposure.
The lifecycle order defines governance relationships.
It must not falsify Operational Reality. Governance Requirements
HCS requires that:
- Harm Context remain distinguishable from Harm itself
- Harm-relevant triggers not automatically establish Harm
- affected subjects and objects be sufficiently identifiable for
the
intended governance purpose; - individual, multiple, collective, systemic,
infrastructural, environmental, and societal contexts remain supportable;
- relevant Harm domains be explicit where they materially affect
governance; - multi-domain and cross-domain Harm contexts remain
governable; - temporal context preserve duration, latency, accumulation,
persistence, recurrence, and uncertainty where relevant; - recognition
time not be substituted automatically for Harm emergence time; - temporal
precision not exceed available support; - operational mechanisms remain
distinguishable from Harm Context; - non-linguistic autonomous action
remain governable as Harm-relevant context; - physical interaction remain
distinguishable from Harm; - continuous, accumulated, latent, and
propagated Harm contexts remain supportable; - material context
relationships remain identifiable without automatically establishing
causation; - causation remain distinct from accountability and liability;
- context uncertainty and competing interpretations remain preservable; -
Context Sufficiency remain distinguishable from Context Completeness; -
downstream conclusions not exceed what the governed context supports; -
material context changes trigger reassessment where necessary; -
historical context remain reconstructable; - updated context not silently
erase prior governed states; - upstream architecture outputs be consumed
without duplication of their governed spaces; - HCS handoff into
downstream HAGA™ governance remain explicit and traceable.
Governance Logic
HCS applies the following Parent Standard-level logic:Has a legitimate Harm-governance trigger arisen?
If no:
→ HCS need not activate.
If yes:
→ identify the Harm-relevant context.
Who or what may be affected?
→ establish the relevant subject, object, group, system, infrastructure,
environment, or population context.
Within which domain or domains may Harm be relevant?
→ establish the applicable Harm Domain Context without assuming Harm
existence.
What temporal context is materially relevant?
→ preserve identifiable occurrence time, duration, accumulation, latency,
persistence, recurrence, or temporal uncertainty.
Is an operational mechanism relevant?
If yes:
→ preserve only the mechanism context required for Harm governance and
retain the appropriate cross-architecture boundary.
Does the context involve physical or autonomous interaction?
If yes:
→ preserve relevant interaction states without equating contact, force,
autonomous action, or AI behavior with Harm.
Are multiple plausible contexts materially possible?
If yes:
→ preserve uncertainty or competing context states rather than
manufacturing certainty.
Is the context sufficient for the intended next governance step?
If no:
→ maintain a Pending, Partial, Disputed, or otherwise appropriately
qualified context state.
If yes:
→ establish Governed Harm Context and permit the appropriate downstream
HAGA™ handoff.
Has material new information changed the context?
If yes:
→ reassess while preserving historical context continuity.
At every stage:
Harm Context ≠ Established Harm
Governance Outputs
HCS may produce:- Governed Harm Context
- Harm Context State
- Harm Governance Trigger Record
- Affected Subject / Object Context
- Collective Harm Context
- Harm Domain Context
- Cross-Domain Harm Context
- Temporal Harm Context
- Latency Context
- Accumulation Context
- Operational Mechanism Context
- Physical Interaction Context
- Autonomous Interaction Context
- Material Context Relationship Record
- Context Uncertainty State
- Competing Context Record
- Context Sufficiency Determination
- Context Reassessment Record
- Historical Harm Context Record
- HAGA™ Lifecycle Handoff State
The principal output is:
Governed Harm Context
capable of supporting the next legitimate stage of HAGA™ governance.Critical Distinctions
Harm Context ≠ HarmHarm Governance Trigger ≠ Harm Emergence
Harm-Relevant Occurrence ≠ Established Harm
Harm-Relevant Signal ≠ Harm Determination
Available Information ≠ Sufficient Harm Context
Context Sufficiency ≠ Context Completeness
Single Affected Subject ≠ Required Harm Structure
Affected Subject Identification ≠ Identity Governance
Harm Domain Context ≠ Harm Classification Finality
Recognition Time ≠ Harm Emergence Time
Observation Time ≠ Harm Origin Time
Operational Mechanism Context ≠ Operational Mechanism Governance
Absence of Linguistic Reasoning ≠ Absence of Harm-Relevant Context
Physical Contact ≠ Harm
Force ≠ Harm
Machine Action ≠ Harm
AI Output ≠ Harm
Unsafe Condition ≠ Established Harm
Context Relationship ≠ Causal Determination
Causal Relevance ≠ Accountability
Accountability ≠ Liability
One Available Narrative ≠ Governed Harm Context
Current Harm Context ≠ Historical Harm Context
Context Change ≠ Historical Context Erasure
Canonical Boundary
HCS governs the identification, formation, structuring, sufficiency,maintenance, reassessment, and historical continuity of the context within
which Harm may become relevant as a distinct governed object. It governs
affected subject and object context, Harm Domain Context, temporal and
latency context, continuous and accumulated context, material
Harm-specific relationships, physical and autonomous interaction context
where relevant, context uncertainty, competing contexts, Context
Sufficiency, material context change, and lifecycle handoff into
downstream HAGA™ governance. It does not establish Harm merely because a
Harm-relevant occurrence, signal, action, interaction, exposure
possibility, allegation, or adverse condition exists; it does not govern
Operational Reality itself, AI behavior, autonomous execution, cognition,
memory, identity, evidence, integrity, general validation, accountability,
legal liability, insurance claims, Value consequences, or Tax-Relevant
Events.
ORA™ and other applicable architectures may establish relevant upstream
governed reality and operational states.
HCS determines what portion of that reality constitutes the governed
context within which Harm must be understood.
HES subsequently governs whether and under what conditions a subject or
object becomes exposed to the possibility of Harm.
Canonical Closing Statement
Harm cannot be governed legitimately when its context is assumed ratherthan established.
The same action, output, contact, condition, failure, or operational event
may be harmless in one context, Harm-relevant in another, and part of
existing Harm in a third. HCS prevents those states from being collapsed
into one another.
HCS establishes the governed context within which the Harm lifecycle can
begin without prematurely declaring that Harm exists, why it occurred, who
is accountable for it, or who may ultimately bear liability.