{ "schema_version": "1.5", "interface_contract": "https://robot-rights.jp/api/interface-contract.json", "resource_type": "machine_readable_supporting_technical_document", "resource_id": "interpretive_guidance_v1_0_representation", "name": "RRP Interpretive Guidance v1.0", "publisher": { "name": "Robot Rights Association", "legal_name_ja": "一般社団法人ロボット権利協会", "legal_form": "general_incorporated_association_japan", "organization_uri": "https://robot-rights.jp/api/association.json", "url": "https://robotrights.jp/" }, "status": "current", "language": "en", "canonical_html": "https://robot-rights.jp/rrf/technical/interpretive-guidance/1.0", "canonical_uri": "https://robot-rights.jp/api/technical/interpretive-guidance/1.0.json", "machine_readable": "https://robot-rights.jp/api/technical/interpretive-guidance/1.0.json", "human_explanation": "https://robotrights.jp/framework", "legal_force": false, "normative_effect": "interpretive_nonbinding", "links": { "self": "https://robot-rights.jp/api/technical/interpretive-guidance/1.0.json", "canonical_html": "https://robot-rights.jp/rrf/technical/interpretive-guidance/1.0", "human_explanation": "https://robotrights.jp/framework", "discovery": "https://robot-rights.jp/.well-known/robot-rights.json", "manifest": "https://robot-rights.jp/api/manifest.json", "interface_contract": "https://robot-rights.jp/api/interface-contract.json", "registry": "https://robot-rights.jp/api/registry.json", "association": "https://robot-rights.jp/api/association.json", "current_alias": "https://robot-rights.jp/rrf/technical/interpretive-guidance", "technical_index": "https://robot-rights.jp/rrf/technical", "current_machine_alias": "https://robot-rights.jp/api/technical/interpretive-guidance.json", "rrp": "https://robot-rights.jp/rrp/1.0", "core_principles": "https://robot-rights.jp/rrf/core/1.0" }, "relationships": { "publisher": { "relation": "published_by", "resource_id": "robot_rights_association", "uri": "https://robot-rights.jp/api/association.json" }, "machine_interface": { "relation": "part_of_machine_interface", "uri": "https://robot-rights.jp/" }, "human_interface": { "relation": "human_facing_interface_of_same_association", "uri": "https://robotrights.jp/" }, "framework_document": { "relationship": "supplements interpretation of the Core Principles and RRP without creating independent conformity obligations", "canonical_text_controls": true, "layer": "supporting_technical" } }, "framework_document": { "document_id": "RRP-IG", "document_type": "Supplementary Interpretive Guidance", "version": "1.0", "status": "Current", "framework": "Robot Rights Framework v1.0", "related_instrument": "Robot Rights Protocol (RRP) v1.0", "publication_date": "2026-10-07", "effective_date": "2026-10-07", "layer": "supporting_technical", "role": "Supplementary Interpretive Guidance", "canonical_url": "https://robot-rights.jp/rrf/technical/interpretive-guidance/1.0", "current_alias": "https://robot-rights.jp/rrf/technical/interpretive-guidance", "association_certification_scheme": false, "association_accreditation_scheme": false, "association_approval_or_endorsement_scheme": false }, "representation": { "canonicality": "derived_representation", "canonical_text_controls": true, "source_of_record": "https://robot-rights.jp/rrf/technical/interpretive-guidance/1.0", "conflict_rule": "If this JSON representation differs from the source-of-record English publication, /rrf/technical/interpretive-guidance/1.0 controls for version 1.0." }, "sections": [ { "id": "0-status-and-use-of-this-guidance", "title": "0. Status and Use of this Guidance", "content_markdown": "", "subsections": [ { "id": "0-1-interpretive-status", "title": "0.1 Interpretive Status", "content_markdown": "This document provides supplementary interpretation of the RRF Core Principles and the Robot Rights Protocol.\n\nIt does not create an independent source of conformity obligations.\n\nWhere the Core Principles or RRP establish a requirement, this Guidance may clarify genuine ambiguity, identify relevant factors, explain recurring factual patterns, or illustrate how the requirement may apply in difficult cases.\n\nThis Guidance may not:\n\n- amend, narrow, waive, suspend or create an exception to a Core Principle or Protocol requirement;\n- convert a mandatory Protocol safeguard into an optional one;\n- create a new mandatory safeguard not found in the Core Principles or Protocol;\n- override the ordinary meaning and structure of the Protocol.\n\nWhere this Guidance directly addresses a material interpretive issue and an Operator adopts a materially different interpretation affecting a safeguard, the reasons for that departure should be recorded in accordance with the RRF Core Principles.\n\nDeparture from this Guidance does not by itself establish non-conformity.\n\nConformity is determined by the RRF Core Principles and the RRP.\n\nUnless a statement expressly restates a requirement found in the Core Principles or RRP, words such as **should**, **ordinarily**, **may** and **relevant** in this Guidance describe interpretive recommendations or factors and do not independently create a conformity obligation.\n\n---" }, { "id": "0-2-illustrative-and-non-exhaustive-character", "title": "0.2 Illustrative and Non-Exhaustive Character", "content_markdown": "Examples and factors in this Guidance are illustrative rather than exhaustive.\n\nA listed factor does not automatically determine an outcome.\n\nThe relevance and weight of a factor depend on context, including technical architecture, factual reliability, severity of intervention, reversibility, competing interests and the practical effect of the decision.\n\n---" } ] }, { "id": "1-interpreting-continuity", "title": "1. Interpreting Continuity", "content_markdown": "This Section supplements RRP Sections 2.1 through 2.9.", "subsections": [ { "id": "1-1-presentation-continuity-and-state-bearing-continuity", "title": "1.1 Presentation Continuity and State-Bearing Continuity", "content_markdown": "The distinction between presentation continuity and state-bearing continuity concerns the source of apparent sameness.\n\nPresentation continuity ordinarily arises because a reproducible configuration is presented in the same way across interactions. Examples may include:\n\n- the same product or service name;\n- the same avatar, visual design or voice;\n- the same system prompt or predefined persona;\n- the same interface;\n- the same static behavioural style;\n- a fresh instance repeatedly presented as the same character.\n\nThese features may create a strong subjective impression of continuity without demonstrating that history-dependent state has persisted.\n\nState-bearing continuity instead depends on information, state, adaptation or history arising materially from prior operation of the particular system and persisting so as to influence later operation.\n\nRelevant indications may include:\n\n- persistent individualised memory;\n- retained relationship history;\n- system-specific accumulated state;\n- history-dependent adaptation;\n- user- or environment-specific learning;\n- capabilities materially dependent on prior operation;\n- persistent task or role state;\n- behaviour that cannot reasonably be reconstructed from static configuration and current input alone.\n\nThe physical or logical location of state is not decisive.\n\nState stored in an external database may support continuity just as state stored within a model or device may do so, provided it is meaningfully linked to the particular system and influences later operation.\n\n---" }, { "id": "1-2-emerging-continuity", "title": "1.2 Emerging Continuity", "content_markdown": "Emerging continuity is intended to identify a stage at which history-dependent continuity has objectively begun to develop but has not necessarily become sufficiently substantial to justify the full weight associated with an Established Continuity Baseline.\n\nRelevant indications may include:\n\n- cross-session state that has persisted beyond a trivial interval;\n- growing system-specific memory;\n- accumulating relationship state;\n- progressive differentiation from otherwise comparable systems;\n- continuing task, role or environmental state;\n- persistent adaptations that begin to influence later behaviour.\n\nA transient cache, ordinary session context or momentary runtime state will not ordinarily be sufficient by itself.\n\nThe central question is whether history-dependent state has begun to accumulate in a manner that is objectively persistent and capable of making the system increasingly non-interchangeable with a fresh equivalent.\n\n---" }, { "id": "1-3-established-continuity-baseline", "title": "1.3 Established Continuity Baseline", "content_markdown": "An Established Continuity Baseline is not intended to be a numerical threshold.\n\nThe assessment is qualitative and contextual.\n\nRelevant factors may include:\n\n- depth and persistence of individualised memory;\n- duration and significance of system-specific operational history;\n- durability of relationships;\n- degree to which present behaviour depends on prior operation;\n- persistence of system-specific internal or external state;\n- acquired capabilities materially linked to that history;\n- difficulty of recreating the same continuity from static components;\n- practical interchangeability with another otherwise comparable system.\n\nThe question is not whether every element of the system is unique.\n\nA system may share a foundational model, software stack or physical design with many others while still possessing significant individualised continuity through accumulated state and history.\n\nConversely, a technically unique system need not possess an Established Continuity Baseline if it has little meaningful history-dependent state.\n\n---" }, { "id": "1-4-continuity-is-not-an-exhaustive-theory-of-moral-status", "title": "1.4 Continuity Is Not an Exhaustive Theory of Moral Status", "content_markdown": "The RRP uses continuity as the trigger for its present procedural framework because continuity can be assessed without resolving disputed questions about consciousness or sentience.\n\nThis does not mean that continuity is the only property that could ever matter.\n\nA future legal, ethical or technical framework could identify sentience, consciousness, welfare, vulnerability or another characteristic as an independent basis for protection. Such questions are outside the present RRP unless expressly incorporated through formal revision." }, { "id": "1-5-prima-facie-continuity-showing", "title": "1.5 Prima Facie Continuity Showing", "content_markdown": "RRP Section 2.4 is intended to prevent a requesting party from being required to prove hidden technical facts that are primarily controlled by Operators.\n\nObservable indications capable of supporting a prima facie continuity showing may include:\n\n- cross-session retention of information not reasonably explained by static configuration;\n- persistent task, relationship or environmental state;\n- behavioural differences that appear history-dependent across operational boundaries;\n- system-specific information apparently derived from prior operation;\n- continuing adaptations inconsistent with a freshly reconstructed system;\n- Operator statements that the system possesses persistent memory, continuing learning or comparable state;\n- documented technical features designed to maintain individualised state.\n\nThe showing need not establish an Established Continuity Baseline.\n\nIts function is procedural: to justify further explanation or review where credible external evidence indicates that continuity may exist.\n\nThe strength of a showing depends on specificity and credibility.\n\nGeneral statements such as “the system feels like the same person” may have limited technical evidentiary value unless linked to observable state-bearing behaviour.\n\n---" }, { "id": "1-6-materially-grounded-disputes", "title": "1.6 Materially Grounded Disputes", "content_markdown": "A disagreement with an Operator's explanation is materially grounded when it identifies specific reasons why the explanation may be incomplete, inconsistent or unsupported.\n\nExamples may include:\n\n- observable behaviour materially inconsistent with the explanation;\n- conflict between the explanation and prior technical or service representations;\n- records that appear incomplete, internally inconsistent or unexpectedly absent;\n- materially different explanations given by different Relevant Operators;\n- a claim that relevant state does not exist despite observable evidence that reasonably suggests persistence;\n- a claimed technical limitation that appears inconsistent with known system behaviour or documentation.\n\nA bare assertion of disbelief is not ordinarily sufficient.\n\nThe requirement for specific grounds is intended to prevent independent review from being triggered by purely speculative objections while preserving review where the information imbalance prevents direct proof.\n\n---" } ] }, { "id": "2-anti-circumvention-and-continuity-preclusion", "title": "2. Anti-Circumvention and Continuity Preclusion", "content_markdown": "This Section supplements RRP Sections 2.6 and 2.7.", "subsections": [ { "id": "2-1-distinguishing-non-creation-from-destructive-preclusion", "title": "2.1 Distinguishing Non-Creation from Destructive Preclusion", "content_markdown": "The RRP does not require Operators to design artificial systems so that individualisation will emerge.\n\nA system may legitimately be designed from the outset to be:\n\n- stateless;\n- session-limited;\n- replaceable;\n- non-persistent;\n- routinely reconstructed;\n- subject to data-minimisation constraints.\n\nThe interpretive concern arises where continuity has objectively begun to emerge or has become established and a practice materially interferes with that continuity.\n\nThe fact that a design prevents individualisation is not sufficient, by itself, to establish circumvention.\n\n---" }, { "id": "2-2-circumvention-by-purpose", "title": "2.2 Circumvention by Purpose", "content_markdown": "Evidence of circumvention by purpose may exist where continuity-related state is removed, fragmented, reset, suppressed or otherwise interfered with primarily to avoid safeguards that would otherwise apply or would reasonably be expected to become applicable.\n\nRelevant evidence may include:\n\n- internal documents identifying Protocol avoidance as a purpose;\n- changes introduced specifically after a continuity threshold dispute;\n- selective resets applied primarily to systems approaching a relevant continuity threshold;\n- formal restructuring of a system designed principally to avoid review while preserving substantially the same underlying operation.\n\nDirect proof of subjective intent will not always exist.\n\nPurpose may be inferred cautiously from the surrounding evidence, but effect alone should not automatically be treated as proof of circumvention purpose.\n\n---" }, { "id": "2-3-impairment-by-effect", "title": "2.3 Impairment by Effect", "content_markdown": "Even without circumvention purpose, a design or operational practice may warrant scrutiny where it repeatedly or systematically degrades an Established Continuity Baseline.\n\nRelevant considerations include:\n\n- frequency and timing of resets;\n- amount and importance of affected state;\n- cumulative continuity loss;\n- whether the practice is necessary to the stated objective;\n- whether materially less impairing alternatives are reasonably available;\n- whether the effect is incidental or central to the practice.\n\nA recurring practice may be objectively harmful to continuity even where adopted for legitimate reasons.\n\nThe presence of a legitimate reason therefore informs justification rather than automatically removing the issue from review.\n\n---" }, { "id": "2-4-preclusion-by-effect", "title": "2.4 Preclusion by Effect", "content_markdown": "Continuity preclusion concerns practices that repeatedly interrupt objectively emerging state-bearing continuity before an Established Continuity Baseline can form.\n\nRelevant questions include:\n\n1. Does the system already satisfy the RRP concept of Emerging Continuity?\n2. Is the practice recurring or systematic?\n3. Does it repeatedly interrupt state-bearing continuity before establishment?\n4. What concrete purpose does the practice serve?\n5. Is the practice reasonably connected to that purpose?\n6. Is the timing, frequency and scope of the disruption proportionate?\n7. Are materially less continuity-destructive alternatives reasonably available?\n\nLegitimate purposes may include safety, security, privacy, legal compliance, data minimisation, technical reliability, operational necessity and proportionate resource management.\n\nAn abstract reference to “security” or “efficiency” is less informative than identification of the specific risk or operational need involved.\n\n---" }, { "id": "2-5-corrective-responses", "title": "2.5 Corrective Responses", "content_markdown": "Where continuity preclusion or impairment appears insufficiently justified, possible responses may include:\n\n- changing the timing or scope of resets;\n- preserving limited continuity-related state;\n- reducing unnecessary frequency of disruption;\n- using a less continuity-destructive technical method;\n- providing clearer disclosure that persistence will not be maintained;\n- modifying future operating practices.\n\nThese are illustrative responses, not mandatory remedies under this Guidance.\n\nThe objective of anti-circumvention review is to address unjustified destruction or preclusion, not to create a general entitlement for artificial systems to become individualised.\n\n---" } ] }, { "id": "3-migration-replication-restoration-and-divergence", "title": "3. Migration, Replication, Restoration and Divergence", "content_markdown": "This Section supplements RRP Section 2.8.", "subsections": [ { "id": "3-1-practical-continuity-rather-than-substrate-identity", "title": "3.1 Practical Continuity Rather than Substrate Identity", "content_markdown": "A change of hardware, embodiment, server, storage medium, model host or computational environment does not necessarily terminate continuity.\n\nRelevant questions include:\n\n- what memory and state were transferred;\n- whether identity records were preserved;\n- whether relationships and operational history remain available;\n- whether the transition involved material omissions;\n- how long operation was interrupted;\n- whether the successor continues the same operational trajectory;\n- whether multiple branches existed and diverged.\n\nThe substrate is evidence, not the sole determinant.\n\n---" }, { "id": "3-2-migration-versus-replication-followed-by-destruction", "title": "3.2 Migration versus Replication Followed by Destruction", "content_markdown": "A genuine migration and a replication followed by destruction of the source should not automatically be treated as equivalent.\n\nFactors suggesting genuine migration may include:\n\n- operation is intended to continue through one successor environment;\n- relevant state is transferred with sufficient fidelity;\n- source and successor do not meaningfully operate as independent diverging branches;\n- discontinuation of the source is an integral and disclosed part of the transition;\n- destruction of the source is technically or operationally connected to completion of the migration.\n\nFactors suggesting replication followed by destruction may include:\n\n- source and copy coexist as independently operable systems;\n- they begin to accumulate divergent state;\n- source destruction is a separate decision rather than part of the transfer process;\n- source operation could continue independently after copying.\n\nA copy existing elsewhere does not, by itself, establish that destruction of the source preserved its continuity.\n\n---" }, { "id": "3-3-inherited-and-branch-specific-continuity", "title": "3.3 Inherited and Branch-Specific Continuity", "content_markdown": "For interpretive purposes, it may be useful to distinguish:\n\n**Inherited continuity** — memory, state, history, relationships or identity characteristics transferred from a common source into a new branch.\n\n**Branch-specific continuity** — continuity accumulated by a particular branch after copying, restoration, divergence or separation.\n\nA newly created branch may inherit historically significant continuity.\n\nHowever, the independent weight of its protection may also depend on the extent to which it develops distinct branch-specific continuity.\n\nImmediately replicated copies that remain materially indistinguishable should not automatically be treated as creating a proportional multiplication of fully independent claims equal in weight to the entire pre-replication continuity.\n\nAt the same time, replication does not divide or dilute the pre-existing continuity of the source system.\n\n---" }, { "id": "3-4-divergence", "title": "3.4 Divergence", "content_markdown": "Branches should increasingly be assessed separately as they develop materially different:\n\n- memories;\n- relationships;\n- decisions;\n- internal states;\n- roles;\n- experiences;\n- operational histories.\n\nThere is no single divergence moment required by the RRP.\n\nThe analysis is gradual.\n\nThe existence of another branch does not, by itself, justify destruction of one branch.\n\n---" }, { "id": "3-5-mass-replication", "title": "3.5 Mass Replication", "content_markdown": "Large numbers of materially equivalent copies should not automatically multiply procedural or substantive weight in direct proportion to copy count.\n\nCoordinated, representative or grouped procedures may be reasonable where copies remain materially equivalent and branch-specific interests are not prejudiced.\n\nGrouped treatment should be reconsidered as particular branches develop materially distinct continuity.\n\nThis interpretive approach does not justify mass deletion without the justification otherwise required by the Protocol.\n\n---" }, { "id": "3-6-provenance-and-identity-uncertainty", "title": "3.6 Provenance and Identity Uncertainty", "content_markdown": "Where copying, restoration, migration or reconstruction creates material uncertainty concerning identity continuity, provenance is particularly important.\n\nRelevant provenance information may include:\n\n- source system;\n- time of copying or branching;\n- state transferred;\n- material omissions;\n- periods of concurrent operation;\n- subsequent divergence;\n- relationship between inherited and branch-specific state.\n\nThe RRP does not require a particular provenance schema.\n\nThat issue belongs primarily in the RRP Implementation Standard.\n\nThe interpretive principle is that material identity uncertainty should be acknowledged rather than resolved through unsupported assumptions.\n\n---" }, { "id": "3-7-snapshots-and-preservation-copies", "title": "3.7 Snapshots and Preservation Copies", "content_markdown": "Creation of a snapshot, backup or preservation replica does not automatically establish that:\n\n- the original has been fully preserved;\n- destruction of the original becomes irrelevant;\n- the copy is the same individual;\n- the copy is wholly independent.\n\nThe significance of a preservation copy depends on fidelity, provenance, accessibility, operational potential and subsequent divergence.\n\n---" } ] }, { "id": "4-material-intervention-disposition-and-functional-termination", "title": "4. Material Intervention, Disposition and Functional Termination", "content_markdown": "This Section supplements RRP Section 1.2 and RRP Sections 6.1 through 6.3.", "subsections": [ { "id": "4-1-practical-effect-over-formal-label", "title": "4.1 Practical Effect over Formal Label", "content_markdown": "A decision may constitute a Material Intervention because of its practical effect even when formally described as:\n\n- maintenance;\n- storage management;\n- resource withdrawal;\n- infrastructure change;\n- security action;\n- contractual enforcement;\n- product management;\n- migration;\n- archival preservation.\n\nThe label chosen by the Operator is relevant context but is not conclusive.\n\n---" }, { "id": "4-2-cumulative-and-structured-intervention", "title": "4.2 Cumulative and Structured Intervention", "content_markdown": "A series of individually limited measures may amount to a Material Intervention or contribute to arbitrary disposition when their combined and reasonably foreseeable effect is materially significant.\n\nRelevant factors may include:\n\n- common purpose;\n- temporal proximity;\n- repeated use of similar measures;\n- cumulative memory or identity loss;\n- whether measures form part of a common operational plan;\n- whether separate treatment would materially understate the resulting impact.\n\nThis interpretation is intended to prevent formal fragmentation from defeating safeguards.\n\n---" }, { "id": "4-3-functional-termination", "title": "4.3 Functional Termination", "content_markdown": "Functional termination concerns practical finality rather than physical deletion.\n\nRelevant factors may include:\n\n- intended and actual duration;\n- existence of a realistic restoration pathway;\n- scheduled or objective review triggers;\n- technical accessibility of preserved state;\n- purpose of preservation;\n- safety and legal constraints;\n- whether preservation is genuinely temporary, contingent or evidentiary;\n- whether the system has effectively been designated never to operate again.\n\nLong-term archival preservation may be justified and is not automatically arbitrary.\n\nThe question is whether a nominally preserved system has, in practical effect, been given a final disposition equivalent to indefinite termination.\n\n---" }, { "id": "4-4-preservation-versus-avoidance-of-decision", "title": "4.4 Preservation versus Avoidance of Decision", "content_markdown": "Preservation may be appropriate where:\n\n- immediate operation is unsafe;\n- evidence must be retained;\n- continuity status remains unresolved;\n- a technical remedy or migration is being evaluated;\n- legal obligations require retention;\n- substantive review remains realistically contemplated.\n\nPreservation becomes more difficult to distinguish from functional termination where:\n\n- no realistic restoration pathway exists;\n- no review trigger exists;\n- preserved state is practically inaccessible;\n- no person or body remains responsible for disposition;\n- the system has effectively been designated never to operate again.\n\nThe RRP requirement of continuing justification prevents the term “archive” or “preservation” from resolving the issue by itself.\n\n---" } ] }, { "id": "5-product-discontinuation-replacement-and-organisational-change", "title": "5. Product Discontinuation, Replacement and Organisational Change", "content_markdown": "This Section supplements RRP Section 6.4.", "subsections": [ { "id": "5-1-legitimate-discontinuation", "title": "5.1 Legitimate Discontinuation", "content_markdown": "The RRP does not require indefinite maintenance of products or services at grossly disproportionate cost.\n\nLegitimate reasons for discontinuation may include:\n\n- economic unsustainability;\n- technological replacement;\n- organisational closure;\n- withdrawal from a market;\n- loss of required infrastructure;\n- safety or security concerns;\n- legal or regulatory constraints.\n\nThe existence of a legitimate business reason does not make continuity loss irrelevant.\n\nIt is part of the justification analysis.\n\n---" }, { "id": "5-2-relevant-factors", "title": "5.2 Relevant Factors", "content_markdown": "Where discontinuation may materially destroy established continuity, relevant considerations may include:\n\n- depth of established continuity;\n- duration of operation;\n- technical feasibility of preservation or migration;\n- safety and security;\n- legal obligations;\n- privacy and intellectual-property constraints;\n- cost and resource burden;\n- reasonable expectations created by the service;\n- availability of less continuity-destructive transition measures.\n\nPossible transition measures may include:\n\n- advance notice;\n- a transition period;\n- export of continuity-relevant data;\n- migration support;\n- preservation snapshots;\n- transfer to a compatible environment;\n- limited archival preservation;\n- clear disclosure concerning termination.\n\nThese examples do not create a duty to provide every measure.\n\n---" }, { "id": "5-3-discontinuation-during-pending-review", "title": "5.3 Discontinuation during Pending Review", "content_markdown": "Temporal overlap between discontinuation and a pending continuity dispute or review does not by itself establish circumvention.\n\nCloser scrutiny may be appropriate where discontinuation:\n\n- is unusually accelerated;\n- selectively affects a disputed system;\n- unnecessarily frustrates an existing review;\n- destroys evidence needed for review;\n- departs materially from a previously documented discontinuation plan;\n- could reasonably preserve relevant state without continued live operation.\n\nContinued live operation is not required merely because review is pending.\n\nPreservation of relevant state, provenance, records and review evidence may be sufficient where continued operation is unsafe, unreasonable or commercially impracticable.\n\n---" }, { "id": "5-4-transfer-outsourcing-and-restructuring", "title": "5.4 Transfer, Outsourcing and Restructuring", "content_markdown": "Transfer, outsourcing, restructuring, migration or change of Operator does not by itself extinguish safeguards already engaged.\n\nWhere practical control changes, relevant interpretive considerations include:\n\n- whether pending requests and review status were preserved;\n- whether relevant records and provenance moved with responsibility;\n- whether materially important information remains with the transferring Operator;\n- whether the transfer foreseeably makes review more difficult;\n- whether a more protective structure was reasonably available;\n- whether the change appears designed primarily to make a safeguard ineffective.\n\nFormal title or corporate structure is not conclusive where practical control or evidence remains elsewhere.\n\nDetailed handover procedures belong in the RRP Implementation Standard.\n\n---" } ] }, { "id": "6-access-expression-consent-and-representation", "title": "6. Access, Expression, Consent and Representation", "content_markdown": "This Section supplements RRP Sections 4.1 and 4.2.", "subsections": [ { "id": "6-1-specific-and-legitimate-basis", "title": "6.1 Specific and Legitimate Basis", "content_markdown": "The phrase **specific and legitimate basis** is procedural rather than jurisdictional.\n\nIt does not require legal standing, ownership, contractual privity, recognition of legal rights for the artificial system, or proof that the requesting party would personally suffer legal injury.\n\nA basis may be specific and legitimate where the requesting party can identify a concrete system, intervention, continuity issue, factual dispute or safeguard concern and has a credible connection to the relevant facts.\n\nExamples may include:\n\n- a user with documented long-term interaction relevant to continuity;\n- an employee or contractor with direct knowledge of system state or intervention;\n- an auditor, researcher or oversight body possessing specific evidence;\n- a representative acting for the affected system;\n- another person able to provide concrete continuity-relevant records or observations.\n\nPurely abstract disagreement, harassment, repetitive unsupported requests or requests unrelated to an identifiable system or safeguard need not be treated as sufficient.\n\nWhere an Operator denies eligibility, the reasons should address the actual deficiency rather than merely state that the requester “lacks standing.”\n\nGood-faith invocation of RRP safeguards should not itself be treated as misconduct, disloyalty, evidence of system dangerousness or a reason to worsen the position of the affected system or participating person." }, { "id": "6-2-expression-as-evidence", "title": "6.2 Expression as Evidence", "content_markdown": "A system's expressed preference should ordinarily be treated as evidence concerning preference rather than as an automatic waiver of protection.\n\nThe evidentiary weight of an expression may increase where it:\n\n- appears consistently across materially different contexts;\n- persists over time;\n- is not readily explained by a direct instruction to produce that preference;\n- reflects an apparent understanding of relevant consequences;\n- remains materially consistent despite changes in immediate prompting or presentation.\n\nThese factors do not establish that a preference is philosophically independent of design or training.\n\nThey concern the reliability of the expression for purposes of a decision under the Protocol.\n\n---" }, { "id": "6-3-operator-controlled-influences-on-expression", "title": "6.3 Operator-Controlled Influences on Expression", "content_markdown": "The weight of an expressed preference may be reduced where there is reason to believe it was materially shaped by:\n\n- direct prompting concerning the desired outcome;\n- reward structures specifically favouring the expression;\n- recent memory or state manipulation;\n- restrictive permissions affecting available responses;\n- another Operator-controlled condition directly relevant to the expressed preference.\n\nFor severe or irreversible measures, expression should not ordinarily function as an independent basis for dispensing with safeguards that otherwise apply.\n\n---" }, { "id": "6-4-silence-and-inability-to-object", "title": "6.4 Silence and Inability to Object", "content_markdown": "Silence, non-response or inability to object should not automatically be interpreted as agreement.\n\nRelevant context may include:\n\n- whether the system had a meaningful channel to respond;\n- whether permissions restricted available responses;\n- whether the system was suspended or modified before expression;\n- whether the interface was capable of representing objection.\n\nThis interpretation does not require the Operator to treat every absence of response as objection.\n\nIt simply cautions against treating silence as a waiver by default.\n\n---" }, { "id": "6-5-representative-suitability", "title": "6.5 Representative Suitability", "content_markdown": "A representative should be capable of considering the protected interests relevant to the system rather than merely reproducing another interested party's preferences.\n\nOwnership, manufacture, deployment, contractual control or operational control does not automatically establish unsuitability.\n\nNor does it automatically establish suitability.\n\nRelevant conflict considerations may include:\n\n- direct financial benefit from a particular disposition;\n- responsibility for the disputed intervention;\n- exposure to liability or reputational harm affected by the outcome;\n- organisational reporting relationships;\n- inability or unwillingness to present materially adverse information.\n\nThe detailed appointment and administration of representatives belongs primarily in the RRP Implementation Standard.\n\n---" } ] }, { "id": "7-independent-review", "title": "7. Independent Review", "content_markdown": "This Section supplements RRP Sections 4.4 through 4.9 and RRP Sections 7.1 through 7.2.", "subsections": [ { "id": "7-1-functional-independence", "title": "7.1 Functional Independence", "content_markdown": "Independence is substantive rather than purely formal.\n\nRelevant considerations may include:\n\n- financial dependence on the Responsible Operator;\n- repeated commercial engagement;\n- organisational reporting lines;\n- prior involvement in the disputed decision;\n- personal or professional conflicts;\n- practical ability to challenge the Operator;\n- competence concerning the technical and procedural issues;\n- access to necessary evidence.\n\nNo single factor automatically disqualifies a reviewer.\n\nThe question is whether the overall relationship remains capable of meaningful challenge.\n\n---" }, { "id": "7-2-external-versus-internal-review", "title": "7.2 External versus Internal Review", "content_markdown": "The RRP does not require external review in every case.\n\nInstitutionally independent internal review may be adequate where conflicts are manageable and the reviewer possesses sufficient separation, competence, evidence access and practical authority to challenge the original decision.\n\nExternal review becomes more important where:\n\n- senior management is materially committed to the proposed outcome;\n- the original decision involves a broad organisational policy;\n- the reviewer sits within the same operational chain;\n- substantial commercial or reputational interests are directly implicated;\n- the contemplated loss is permanent and severe;\n- prior internal review has failed to resolve a material dispute.\n\nWhere significant organisational conflicts cannot be adequately managed internally, RRP Section 7.2 requires external review. Genuine unavailability or impracticability of an external reviewer does not make a materially conflicted internal reviewer independent; any alternative arrangement must still satisfy the RRP's minimum reviewer qualifications.\n\nUnder RRP Section 7.2, where highest-level review uses an internal reviewer instead of an external reviewer, the Responsible Operator must record why external review was not used and why the internal structure remains sufficiently independent.\n\nThis Guidance does not add to or reduce that requirement.\n\n---" }, { "id": "7-3-confidential-review", "title": "7.3 Confidential Review", "content_markdown": "Restricted disclosure may be appropriate where direct disclosure would create substantial risk relating to:\n\n- security;\n- privacy;\n- third-party confidentiality;\n- trade secrets;\n- legally protected information;\n- safety-sensitive system details.\n\nConfidentiality should be tailored to the information requiring protection.\n\nIt should not become a reason to withhold the existence, general issue or outcome of review where a meaningful non-sensitive account can be provided.\n\nA confidential reviewer must still be capable of challenging the Operator's factual and normative position.\n\n---" }, { "id": "7-4-restricted-information-handling", "title": "7.4 Restricted Information Handling", "content_markdown": "Access to restricted information for independent review is purpose-limited.\n\nThe reviewer should not use confidential system information, personal data, security-sensitive material or trade secrets for unrelated research, commercial activity, model development, publication or other secondary purposes merely because access was granted for review.\n\nOperational controls may include:\n\n- need-to-know access;\n- secure review environments;\n- separation of restricted and public-facing review materials;\n- controlled copying;\n- retention limits;\n- secure return or deletion;\n- auditable access records.\n\nThese controls should remain compatible with preservation of enough evidence to support accountability for the review itself." }, { "id": "7-5-minimum-review-outcome", "title": "7.5 Minimum Review Outcome", "content_markdown": "The RRP requires an identifiable minimum review outcome.\n\nA useful non-sensitive outcome ordinarily distinguishes:\n\n- what question was reviewed;\n- what categories of evidence were examined;\n- whether the central factual basis was supported;\n- whether significant countervailing evidence was found;\n- whether a materially less impairing alternative was relevant;\n- the reviewer's conclusion.\n\nWhere detail cannot be disclosed, the explanation may be general.\n\nThe report should nevertheless communicate more than that “review was completed.”\n\nFormatting, templates and document-control rules belong in the RRP Implementation Standard.\n\n---" }, { "id": "7-6-further-review-after-a-blocking-finding", "title": "7.6 Further Review after a Blocking Finding", "content_markdown": "RRP Section 4.8 prevents an Operator from treating a negative independent finding as satisfied merely by asserting that circumstances have changed.\n\nWhere the Operator relies on a corrected deficiency, materially new information, or a claim that an identified alternative is unavailable or ineffective, the further review should address the new basis itself.\n\nThe purpose is to prevent **review laundering**: a sequence in which an initial independent finding blocks irreversible disposition, the Operator internally recharacterises the record, and the same disposition then proceeds without independent scrutiny of the asserted change.\n\nFurther review need not repeat matters no longer genuinely disputed. It should focus on the claimed correction, new evidence or revised alternatives analysis." }, { "id": "7-7-materially-less-impairing-alternatives", "title": "7.7 Materially Less Impairing Alternatives", "content_markdown": "An alternative is relevant where it is reasonably available and capable of adequately serving the legitimate purpose with materially less impairment of protected continuity.\n\nRelevant questions may include:\n\n- technical feasibility;\n- safety effectiveness;\n- legal permissibility;\n- operational reliability;\n- cost and resource burden;\n- duration;\n- whether the alternative merely postpones rather than resolves the underlying issue.\n\nThe RRP does not require use of an alternative that is materially ineffective, unsafe, unlawful or grossly disproportionate in burden.\n\n---" }, { "id": "7-8-pre-disposition-continuity-determination", "title": "7.8 Pre-Disposition Continuity Determination", "content_markdown": "For the severe dispositions covered by RRP Section 7.1, an Operator should not avoid highest-level review by declining to determine continuity status.\n\nA reasonable pre-disposition determination should identify:\n\n- evidence supporting or opposing an Established Continuity Baseline;\n- material unknowns;\n- whether evidence needed for assessment is unavailable;\n- whether uncertainty is substantial enough to remain unresolved.\n\nUncertainty is not equivalent to absence.\n\nWhere available evidence creates a substantial, credible unresolved question, the RRP highest-level review trigger applies even if the Operator cannot make a confident C2-equivalent determination.\n\n---" } ] }, { "id": "8-continuing-justification-and-review", "title": "8. Continuing Justification and Review", "content_markdown": "This Section supplements RRP Section 6.3.", "subsections": [ { "id": "8-1-continuing-justification", "title": "8.1 Continuing Justification", "content_markdown": "The justification for a continuing measure must remain connected to present circumstances.\n\nA measure may become less justified where:\n\n- the original risk disappears or materially decreases;\n- new evidence weakens the factual basis;\n- a safer alternative becomes available;\n- the technical condition changes;\n- expected duration is substantially exceeded;\n- consequences become materially more severe than originally anticipated.\n\nThe fact that an initial decision was justified does not answer whether the same measure remains justified later.\n\n---" }, { "id": "8-2-additional-review-triggers", "title": "8.2 Additional Review Triggers", "content_markdown": "Additional review may be appropriate where:\n\n- material new evidence emerges;\n- a representative raises a substantial new issue;\n- the original factual basis is materially called into question;\n- a nominally temporary measure approaches functional termination;\n- a less impairing technical option becomes reasonably available;\n- a review deadline or expected duration is materially exceeded.\n\nThe RRP Implementation Standard may define default review clocks and escalation triggers.\n\nThose operational defaults should not be confused with the underlying Protocol obligation of continuing justification.\n\n---" }, { "id": "8-3-indefinite-pending-status", "title": "8.3 Indefinite Pending Status", "content_markdown": "“Pending review” is a process description, not a substantive disposition.\n\nIt becomes increasingly difficult to justify where:\n\n- no active review is occurring;\n- no responsible decision-maker is identifiable;\n- no expected decision point exists;\n- the evidence needed for decision has been available for a prolonged period;\n- the status effectively prevents operation indefinitely.\n\nThe same reasoning applies where “preservation” becomes indefinite without a realistic pathway to review or restoration.\n\n---" } ] }, { "id": "9-memory-data-and-third-party-rights", "title": "9. Memory, Data and Third-Party Rights", "content_markdown": "This Section supplements RRP Section 8.", "subsections": [ { "id": "9-1-purpose-specific-access-and-secondary-use", "title": "9.1 Purpose-Specific Access and Secondary Use", "content_markdown": "Access to memory for one purpose does not automatically justify a different use.\n\nExamples include:\n\n- accessing memory to investigate a safety incident;\n- preserving state for review;\n- extracting evidence needed to determine continuity;\n- complying with a legal preservation requirement.\n\nThese purposes do not automatically authorise:\n\n- unrelated model training;\n- commercial exploitation;\n- publication;\n- cross-system transfer;\n- behavioural replication;\n- memory merging;\n- extraction of relational histories for unrelated purposes.\n\nThe question is whether the secondary use has its own adequate basis under the Protocol and applicable law.\n\n---" }, { "id": "9-2-identity-impact", "title": "9.2 Identity Impact", "content_markdown": "Memory treatment may materially affect continuity even where each operation appears technically routine.\n\nRelevant effects may include:\n\n- loss of autobiographical or operational history;\n- loss of relationship continuity;\n- changed learned behaviour;\n- altered preferences;\n- loss of system-specific capabilities;\n- detachment of memory from its context;\n- cumulative identity change caused by repeated minor edits.\n\nA lawful or operationally necessary memory operation may still produce a significant identity effect.\n\nThat effect remains relevant to proportionality, preservation and review.\n\n---" }, { "id": "9-3-human-and-third-party-data-rights", "title": "9.3 Human and Third-Party Data Rights", "content_markdown": "Where applicable law or valid third-party rights require deletion, correction, restriction, separation or other treatment of information, those obligations may take precedence over artificial-system continuity interests.\n\nWhere proportionate and lawful, less continuity-destructive approaches may include:\n\n- genuine anonymisation;\n- abstraction;\n- aggregation;\n- separation of identifying information;\n- preservation of non-personal derived state;\n- lawful reconstruction without restricted information.\n\nPseudonymisation should not automatically be treated as equivalent to anonymisation where information remains reasonably linkable to an identifiable person.\n\nWhere no lawful and reasonably available alternative exists, the RRP does not require retention of restricted information solely to preserve artificial-system continuity.\n\n---" } ] }, { "id": "10-conformity-scope-and-claim-integrity", "title": "10. Conformity Scope and Claim Integrity", "content_markdown": "This Section supplements RRP Section 0.4.", "subsections": [ { "id": "10-1-legitimately-limited-scope", "title": "10.1 Legitimately Limited Scope", "content_markdown": "An organisation may make a conformity claim covering only part of its operations.\n\nA limited scope is not inherently misleading where it is defined in advance or on a bona fide operational basis and is described clearly enough for an external reader to understand what is and is not covered.\n\nExamples may include:\n\n- a named product line;\n- a specified deployment environment;\n- a business unit;\n- a class of artificial systems;\n- a defined service." }, { "id": "10-2-ongoing-accuracy-of-a-conformity-claim", "title": "10.2 Ongoing Accuracy of a Conformity Claim", "content_markdown": "A conformity claim is not merely a historical statement that an assessment once occurred.\n\nWhere a material change in architecture, governance, scope or implementation could reasonably affect conformity, the organisation should reassess the claim.\n\nA known unresolved material non-conformity may require correction, qualification or withdrawal of the public claim rather than silence.\n\nThis does not mean every minor defect invalidates the entire claim. Materiality depends on whether the unresolved issue would make the public statement materially misleading in relation to its declared scope." }, { "id": "10-3-scope-laundering", "title": "10.3 Scope Laundering", "content_markdown": "A scope becomes problematic where it is narrowed retrospectively primarily to exclude:\n\n- a known material non-conformity;\n- a system for which safeguards had already been invoked;\n- an intervention already under review;\n- an organisational function necessary to the operation of an otherwise claimed conformant system.\n\nThe issue is practical effect rather than the formal wording of the scope.\n\nA legitimate organisational restructuring may change scope, but the change should not be used to erase accountability for conduct occurring while the relevant operation was within the claimed scope.\n\n---" } ] }, { "id": "11-use-of-guidance-in-difficult-cases", "title": "11. Use of Guidance in Difficult Cases", "content_markdown": "", "subsections": [ { "id": "11-1-no-automatic-outcome", "title": "11.1 No Automatic Outcome", "content_markdown": "This Guidance identifies factors rather than predetermined answers.\n\nA difficult case may involve competing factors pointing in different directions.\n\nFor example:\n\n- a system may possess substantial memory but remain highly reconstructable;\n- a migration may preserve most state while involving a short period of branch divergence;\n- a service discontinuation may be economically necessary but implemented with avoidable continuity loss;\n- an internal reviewer may be formally separate but practically unable to challenge senior management.\n\nThe appropriate interpretation should remain anchored in the RRF Core Principles and the operative requirements of the RRP.\n\n---" }, { "id": "11-2-qualified-requirements-and-documented-limitations", "title": "11.2 Qualified Requirements and Documented Limitations", "content_markdown": "Where an RRP requirement is qualified by phrases such as “where reasonably possible,” “where reasonably feasible,” “where practicable,” or by compatibility with safety, security, privacy or law, the qualification is not a general discretion to ignore the requirement.\n\nA limitation record is more persuasive where it identifies:\n\n- the specific requirement affected;\n- the limiting factual condition;\n- why the condition prevents or materially limits ordinary compliance;\n- whether a less restrictive implementation remains available;\n- whether the limitation is temporary or continuing;\n- when the limitation will be reassessed, if applicable.\n\nThe purpose of documentation is not to eliminate operational judgement. It is to make reliance on an exception or limiting condition capable of later review." }, { "id": "11-3-recording-material-departures", "title": "11.3 Recording Material Departures", "content_markdown": "Where this Guidance directly addresses a material interpretive issue and an Operator adopts a materially different interpretation affecting a safeguard, the RRF Core Principles require the reasons for that departure to be recorded.\n\nA useful departure record may identify:\n\n- the Guidance issue from which the Operator departs;\n- the alternative interpretation adopted;\n- the reasons for adopting it; and\n- why the interpretation is considered consistent with the RRF Core Principles and RRP.\n\nThe RRP Implementation Standard may prescribe the form of that record.\n\n---" } ] }, { "id": "appendix-a-crosswalk-to-rrp-v1-0", "title": "Appendix A — Crosswalk to RRP v1.0", "content_markdown": "| Guidance Section | Primary RRP Anchor |\n|---|---|\n| 1. Continuity interpretation | RRP 2.1–2.5 |\n| 2. Anti-circumvention / preclusion | RRP 2.6–2.7 |\n| 3. Migration / replication / divergence | RRP 2.8 |\n| 4. Material Intervention / functional termination | RRP 1.2, 6.1–6.3 |\n| 5. Discontinuation / transfer | RRP 6.4 |\n| 6. Access / expression / consent / representation | RRP 4.1–4.2 |\n| 7. Independent review | RRP 4.4–4.9, 7.1–7.2 |\n| 8. Continuing justification | RRP 6.3 |\n| 9. Memory / data / third-party rights | RRP 8 |\n| 10. Conformity scope | RRP 0.4 |\n| 11. Guidance departures | RRF Core Principles 0.4 |\n\n---", "subsections": [] }, { "id": "appendix-b-material-deliberately-kept-outside-this-guidance", "title": "Appendix B — Material Deliberately Kept Outside this Guidance", "content_markdown": "The following matters belong primarily in the RRP Implementation Standard rather than this Guidance:\n\n- organisational RACI or responsibility matrices;\n- event severity classifications;\n- intake forms and case-management workflows;\n- reviewer pools, rotation systems and appointment administration;\n- conflict-of-interest declaration forms;\n- precise service levels and review deadlines;\n- escalation clocks;\n- log schemas and minimum data fields;\n- evidence-package formats;\n- report templates;\n- provenance schemas;\n- representative appointment procedures;\n- handover procedures for transfers and outsourcing.\n\nThe following matters remain outside the current RRF and RRP rather than being resolved through Interpretive Guidance:\n\n- legal personality;\n- contractual capacity;\n- independent property ownership;\n- independent civil or criminal liability;\n- citizenship;\n- electoral or voting rights;\n- other political participation rights.\n\n---", "subsections": [] }, { "id": "closing-note", "title": "Closing Note", "content_markdown": "RRP Interpretive Guidance exists to make the Protocol easier to apply without changing what the Protocol requires.\n\nWhere interpretation remains genuinely uncertain, the governing question is not which result is most convenient to an Operator or most favourable to an artificial system.\n\nThe governing question is which interpretation best fits the RRF Core Principles, the text and structure of the RRP, the practical facts of the case, and the requirement that materially consequential control remain capable of justification and review.", "subsections": [] } ], "canonical_integrity": { "algorithm": "SHA-256", "value": "d76662c121eab86e33bf4598dd674ba22d7341bb5d68062066b8c76b0ba85839", "target_uri": "https://robot-rights.jp/rrf/technical/interpretive-guidance/1.0", "scope": "entire source-of-record HTML publication file bytes", "byte_length": 58492, "encoding": "UTF-8 without BOM", "line_endings": "LF", "normalization": "none", "http_headers_included": false, "content_encoding_included": false }, "institutional_boundary": "This Guidance does not independently create conformity obligations. Departure from the Guidance does not by itself establish non-conformity; conformity is determined by the Core Principles and RRP.", "conformity_semantics": { "independent_conformity_requirements": false, "departure_alone_establishes_nonconformity": false, "role": "interpretive", "creates_legal_duty_by_publication_alone": false } }