ready
direct-claim-conflictDetect when two or more sources make mutually incompatible claims about the same entity, number, event, date, or outcome.
Do not smooth over direct contradiction just to produce a cleaner answer.DISCOVERY CONFLICT DETECTION READINESS LAYER
Contradiction is not noise. It is intelligence waiting to be mapped.
A read-only conflict detection readiness layer that defines how TheoB will preserve disagreement, score uncertainty, and map conflicting claims before Vault ingestion or capsule compression.
Universal Time Scroll
Every TheoB pathway can move through Past, Present, and Future without losing context.
Present
Read current signals, conditions, and live context.
🎙
Universal Voice Orb
Voice ready
SYSTEM STATE
StableFounder ControlledHuman Reviewed
CONFLICT DETECTION READINESS
Don’t bury the contradiction. Map it.
Discovery Conflict Detection Readiness prepares TheoB to detect direct claim conflicts, timeline conflicts, measurement conflicts, definition conflicts, methodology conflicts, multimodal conflicts, and cultural-contextual conflicts without pretending uncertainty is failure.
trueConflict Readiness
falseLive Conflict Detection
8Conflict Types
falseAuto Truth Verdict
Conflict detection readiness is active as a non-destructive policy layer. TheoB can define conflict types, severity bands, and future conflict maps, but it cannot compare live provider results, declare final truth, write Vault records, or compress capsules yet.
ready
date-or-timeline-conflictDetect conflicting publication dates, event dates, timelines, sequence claims, or version histories.
Timeline conflicts need source dates, retrieval dates, and update history.foundation-ready
measurement-or-statistic-conflictDetect conflicting measurements, percentages, counts, prices, rankings, scientific values, or dataset-derived statistics.
Units, methodology, region, and reporting period must travel with the number.foundation-ready
definition-or-taxonomy-conflictDetect when sources define the same entity, category, process, or term differently.
Different definitions may reflect context, not error.foundation-ready
source-authority-conflictDetect when lower-authority sources contradict higher-authority or primary sources.
Authority can guide review, but cannot automatically silence minority evidence.review-required
methodology-conflictDetect when sources reach different conclusions because of different methods, samples, instruments, models, or assumptions.
Method conflict requires careful explanation, not winner-take-all scoring.review-required
multimodal-observation-conflictDetect when text claims conflict with images, diagrams, maps, CAD, schematics, datasets, or visual evidence.
Visual evidence must preserve uncertainty zones and interpretation boundaries.review-required
cultural-or-contextual-conflictDetect when cultural, ceremonial, historical, indigenous, scientific, or commercial perspectives frame the same topic differently.
Do not flatten plural perspectives into one false certainty.ready
noneNo meaningful conflict detected among compared sources.
Proceed with normal source scoring and citation trail.ready
minorSmall wording, timing, rounding, or framing differences exist but do not change the core claim.
Mention nuance only when relevant.ready
moderateSources disagree on meaningful details that may affect interpretation.
Surface disagreement and require evidence weighting.foundation-ready
strongSources make incompatible claims that cannot both be true under the same context.
Show conflict map and avoid final certainty.review-required
criticalConflict affects safety, legality, financial impact, medical/health claims, production decisions, or founder-authorized operations.
Require human review before action or Vault ingestion.ready
Conflict Must Be PreservedTheoB must retain disagreement instead of hiding it behind a single averaged answer.
Truth pursuit includes uncertainty.ready
Conflict Is Not FailureA conflict can reveal useful nuance, changing evidence, regional context, or different methodologies.
Do not treat every disagreement as bad data.ready
Claim-Level Conflict RequiredConflict should be detected at the claim level, not only at the source or domain level.
One source can be accurate on one claim and weak on another.ready
Primary Sources Get Strong WeightPrimary-source and high-authority evidence should influence conflict review more than copied summaries.
Authority guides review but does not erase visible dissent.ready
Deduplication Before ConflictCopied repetition must be clustered before conflict is measured.
Do not count syndicated duplicates as independent agreement.ready
No Live Conflict Detection YetThis layer defines readiness only and does not compare live provider results.
No provider queries, Vault writes, or automatic conclusions.ready
No Automatic Truth VerdictConflict detection can identify disagreement but cannot declare final truth by itself.
Conflict maps inform review; they do not replace judgment.review-required
Critical Conflict Requires ReviewHigh-impact conflicts require human review before action, Vault ingestion, or capsule compression.
Do not automate high-stakes decisions from unresolved evidence.review-required
Multimodal Conflict Needs Separate MethodsText, images, diagrams, maps, schematics, CAD, and datasets need specialized cross-modal conflict logic.
Visual contradiction cannot be handled with text-only rules.review-required
Capsules Must Preserve ConflictFuture Intelligence Capsules must retain conflict status, uncertainty, and source trails.
Compression must reduce size, not truth.Allowed NowRender conflict detection readiness.Define conflict types.Define conflict severity bands.Define future conflict map shape.Define future conflict claim shape.Preserve uncertainty as useful intelligence.Keep live conflict detection disabled.Keep automatic truth verdicts disabled.
Not Allowed YetCompare live provider results.Query discovery providers.Declare final truth automatically.Hide unresolved conflicts.Write conflict maps to the Vault.Compress unresolved conflicts into capsules.Use text-only conflict logic for images, CAD, schematics, maps, or datasets.Trigger agent actions from unresolved conflict maps.
conflictMapId: stable conflict map idqueryId: linked discovery query idclaimId: linked claim idconflictType: direct-claim/date-or-timeline/measurement-or-statistic/definition-or-taxonomy/source-authority/methodology/multimodal/cultural-contextualseverity: none/minor/moderate/strong/criticalcomparedReferenceIds: array of compared reference idssupportingReferenceIds: array of references supporting the claimcontradictingReferenceIds: array of references contradicting the claimuncertainReferenceIds: array of references needing reviewduplicateClusterIds: array of duplicate clusters consideredstrongestSourceId: reference id with strongest current support, if anyexplanation: human-readable conflict explanationresolutionStatus: unresolved/partially-resolved/resolved/review-requiredhumanReviewRequired: true/falsecapsulePreservationRequired: trueproductionMutation: false
claimId: stable claim idclaimText: safe claim summarynormalizedClaim: normalized claim form for comparisonentities: array of safe entity labelsvalues: numbers, dates, units, or categorical valuescontext: region, time, domain, method, or source contextsourceTrail: linked reference idsconfidenceBeforeConflict: 0-100confidenceAfterConflict: 0-100uncertainty: low/medium/highevidenceNotes: safe explanation notes
Discovery Vault Ingestion ReadinessTheoB Intelligence Capsule Engine FoundationMultimodal Conflict Detection ReadinessImage Capsule Interpretation LayerDiagram And Schematic Capsule ReadinessVisual Semantics Color Intelligence Registry