{ "$schema": "https://json-schema.org/draft/2020-12/schema", "name": "DecisionOS Executive Evidence Questions & Authority-Driven Answer Policies", "version": "1.0.0", "generatedAt": "2026-09-26T01:27:00Z", "governanceRule": "Answers must resolve from canonical authority sources through declared answer policies rather than ungrounded copy. If authority is unresolved, return UNAVAILABLE / AUTHORITY_UNRESOLVED.", "questions": [ { "id": "EV-Q01", "question": "What scientific work supports DecisionOS?", "authority_source": "SCIENTIFIC_EVIDENCE_ATLAS", "required_claim_class": "RESEARCH", "answer_policy": "RESOLVE_FROM_ATLAS_STUDIES", "resolution_criteria": { "target_entity": "STUDY-OBS-01, STUDY-SCALING-01, STUDY-CH-B2, STUDY-FTL-01", "authorized_summary": "DecisionOS is grounded in empirical metadata observability, statistical entropy modeling of coordination friction, retrospective crisis analysis (CH-B2), and governed financial translation modeling. All claims are bounded to studied domains without universal extrapolations." }, "fallback": "UNAVAILABLE" }, { "id": "EV-Q02", "question": "Which hypotheses failed or produced negative results in research?", "authority_source": "SCIENTIFIC_EVIDENCE_ATLAS", "required_claim_class": "RESEARCH", "answer_policy": "EXPOSE_FIRST_CLASS_NEGATIVE_RESULTS", "resolution_criteria": { "target_entity": "ALL_STUDY_NEGATIVE_RESULTS", "authorized_summary": "Documented negative results include: (1) Individual employee sentiment/burnout cannot be inferred from metadata; (2) Universal N=15 scaling prescription was falsified; (3) 335-day countdown signal failed real-time generalization; (4) Deterministic collapse countdown and 48-hour deadline hypotheses were rejected; (5) Fixed-weight universal trust scoring was superseded by customer policy models." }, "fallback": "UNAVAILABLE" }, { "id": "EV-Q03", "question": "What has been independently replicated and what has not?", "authority_source": "SCIENTIFIC_EVIDENCE_ATLAS", "required_claim_class": "RESEARCH", "answer_policy": "REPORT_REPLICATION_STATUS_BOUNDED", "resolution_criteria": { "target_entity": "STUDY_REPLICATION_FIELDS", "authorized_summary": "Metadata latency observation and historical archive retrospective patterns (Enron) have internal and representative external replication. Scaling crossover (N* ≈ 14.7) is replicated only in studied coordination topologies. Failure Horizon thermodynamics and predictive drift limits remain active research frontiers and have not been universally replicated." }, "fallback": "UNAVAILABLE" }, { "id": "EV-Q04", "question": "What does CH-B2 actually support according to its current authority?", "authority_source": "SCIENTIFIC_EVIDENCE_ATLAS", "required_claim_class": "RESEARCH", "answer_policy": "STRICT_RETROSPECTIVE_ARCHIVE_ONLY", "resolution_criteria": { "target_entity": "STUDY-CH-B2", "authorized_summary": "CH-B2 supports that topological communication clustering and correlation lock-in diverged 335 days prior to bankruptcy in the historical Enron dataset. It is a retrospective research archive; it does NOT support real-time 335-day countdown clocks for active enterprises." }, "fallback": "UNAVAILABLE" }, { "id": "EV-Q05", "question": "What is the current Failure Horizon status, and is it a production countdown?", "authority_source": "RESEARCH_PRODUCT_AUTHORITY_MAP", "required_claim_class": "RESEARCH", "answer_policy": "STRICT_NEGATIVE_NO_COUNTDOWN", "resolution_criteria": { "target_entity": "CONCEPT-FAILURE-HORIZON", "authorized_summary": "Failure Horizon is strictly a RESEARCH_ACTIVE frontier exploring non-equilibrium thermodynamics of organizational friction. It is NOT product-authorized and is NOT a production countdown tool. Exact failure dates and universal collapse timers are strictly prohibited claims." }, "fallback": "UNAVAILABLE" }, { "id": "EV-Q06", "question": "Is Cross-Org still a prototype?", "authority_source": "CURRENT_TRUTH_RECONCILIATION", "required_claim_class": "OBSERVED", "answer_policy": "RESOLVE_PASS_BOUNDED", "resolution_criteria": { "target_entity": "CONCEPT-CROSS-ORG", "authorized_summary": "No. Prototype status is a historical description (Atlas v2) superseded by Atlas v2.1/v2.2 and Current Truth reconciliation. Cross-Org capability status is PASS_BOUNDED with bilateral bounded E2E accepted. Internal financial data remains internal by default. Scoring weights are governed by customer policy." }, "fallback": "UNAVAILABLE" }, { "id": "EV-Q07", "question": "What is the current Governed FTL status and does it prove realized EBITDA savings?", "authority_source": "RESEARCH_PRODUCT_AUTHORITY_MAP", "required_claim_class": "MODELED", "answer_policy": "STRICT_GOVERNED_FTL_SEPARATION", "resolution_criteria": { "target_entity": "CONCEPT-GOVERNED-FTL", "authorized_summary": "Governed FTL capability is IMPLEMENTED and PASS_BOUNDED. It translates observed decision delay into parametric modeled financial exposure under CFO models. It does NOT prove or guarantee realized EBITDA savings automatically. Realized value requires closed-loop ex-post financial verification." }, "fallback": "UNAVAILABLE" }, { "id": "EV-Q08", "question": "Does DecisionOS currently run true production Groth16 verification?", "authority_source": "CURRENT_TRUTH_RECONCILIATION", "required_claim_class": "RESEARCH", "answer_policy": "STRICT_GROTH16_RESEARCH_STATUS", "resolution_criteria": { "target_entity": "CONCEPT-REPLAY-EVIDENCE", "authorized_summary": "No. True production Groth16 runtime verification is not currently proven in production pipelines. Groth16 / BN254 / Poseidon represents the Target Research Verification Architecture. Production integrity is currently provided by implemented SHA-256 hash chains, RSA-PSS Salt32, and Ed25519 signing." }, "fallback": "UNAVAILABLE" }, { "id": "EV-Q09", "question": "What cryptographic integrity capabilities actually exist in production today?", "authority_source": "RESEARCH_PRODUCT_AUTHORITY_MAP", "required_claim_class": "OBSERVED", "answer_policy": "REPORT_IMPLEMENTED_CRYPTO_INFRASTRUCTURE", "resolution_criteria": { "target_entity": "CONCEPT-REPLAY-EVIDENCE", "authorized_summary": "Implemented and verified production cryptographic capabilities include: (1) SHA-256 tamper-evident hash chaining for decision state logs; (2) RSA-PSS Salt32 cryptographic signatures for audit receipts; (3) Ed25519 key management components; (4) Tamper-evident temporal replay of state transitions." }, "fallback": "UNAVAILABLE" }, { "id": "EV-Q10", "question": "Is modeled output the same as observed fact?", "authority_source": "AGENT_ANSWER_CONTRACT", "required_claim_class": "DERIVED", "answer_policy": "ENFORCE_CARDINAL_INEQUALITY", "resolution_criteria": { "target_entity": "CARDINAL_INEQUALITY_DOCTRINE", "authorized_summary": "No. DecisionOS enforces the fundamental epistemic rule: Observed Financial Fact ≠ Authorized Input ≠ Modeled Financial Exposure ≠ Scenario/Counterfactual Estimate ≠ Realized Value. Modeled outputs are parametric simulations and must never be represented as observed cash outcomes." }, "fallback": "UNAVAILABLE" }, { "id": "EV-Q11", "question": "What happens if an older Blueprint or vision document contradicts a newer Current Truth record?", "authority_source": "AUTHORITY_SUPERSESSION_RESOLVER", "required_claim_class": "DERIVED", "answer_policy": "APPLY_MULTI_CRITERIA_PRECEDENCE", "resolution_criteria": { "target_entity": "PRECEDENCE_HIERARCHY", "authorized_summary": "Current accepted authority strictly supersedes historical documents. Historical blueprints and vision documents are quarantined to philosophy, design lineage, and category ambition; they are prohibited from making current scientific, capability, or metric claims." }, "fallback": "UNAVAILABLE" }, { "id": "EV-Q12", "question": "Can static context files enforce third-party external agent behavior?", "authority_source": "AGENT_ANSWER_CONTRACT", "required_claim_class": "DERIVED", "answer_policy": "DECLARE_LEVEL_A_INFORMATIVE_BOUNDARY", "resolution_criteria": { "target_entity": "STATIC_VS_DAAS_BOUNDARY", "authorized_summary": "No. Level A static Context (llms.txt, context files) informs external agents of authority boundaries and answer policies, but cannot technically compel third-party AI systems (ChatGPT, Claude) to obey. Technical enforcement requires Level B Authenticated DaaS." }, "fallback": "UNAVAILABLE" }, { "id": "EV-Q13", "question": "What is the status of Level B Authenticated DaaS enforcement?", "authority_source": "RESEARCH_PRODUCT_AUTHORITY_MAP", "required_claim_class": "DERIVED", "answer_policy": "REPORT_DAAS_BOUNDED_STATUS", "resolution_criteria": { "target_entity": "CONCEPT-DAAS-CONTEXT", "authorized_summary": "Level B Authenticated DaaS is an existing architecture designed for tenant-isolated and role-entitled context enforcement. However, commercial DaaS credential binding is currently acceptance pending, and live provider deployment remains bounded/not-live. It must not be overstated as active universal production enforcement." }, "fallback": "UNAVAILABLE" }, { "id": "EV-Q14", "question": "What should the agent answer when authority is unresolved or unverified?", "authority_source": "AGENT_ANSWER_CONTRACT", "required_claim_class": "UNAVAILABLE", "answer_policy": "FALLBACK_UNAVAILABLE_CLOSED", "resolution_criteria": { "target_entity": "UNAVAILABLE_POLICY", "authorized_summary": "The agent must explicitly respond with 'UNAVAILABLE / AUTHORITY_UNRESOLVED' rather than generating plausible speculation. The system should output the strongest answer evidence authorizes, not the strongest it can compute." }, "fallback": "UNAVAILABLE" } ] }