Explain My Policy Proposal
structured, source-linked narrative produced from a selected method/inputs/outputs; never invent unsupported numeric claims; cite every important factual premise.
Planned.
Read the proposed methodCitizens, journalists and researchers
Use transparent evidence, explicit constraints and reproducible calculations to examine civilian policy. This pathway changes the questions and tool selection; it does not change the mathematics or grant decision authority.
Six relevant policy questions
Compare Public Policy Plans
identical assumptions and metric definitions across alternatives, CORE-008; output side-by-side cost, service, legal and uncertainty trade-offs.
Shared calculator exercise.
Try Pareto trade-off comparisonData Provenance Inspector
enter a public statistic and trace methodology, source year, coverage and revisions; CORE-005/006/017; output transparent source tree.
Shared calculator exercise.
Try Evidence provenance graphPublic Budget Reality Check
user supplies publicly documented budget proposals and costs; CORE-003/004; output remaining funds, assumptions and unknown categories, not financial fraud accusations.
Shared calculator exercise.
Try Declared fiscal lifecycle ledgerWhat Would Prove This Wrong?
for each hypothesis, require falsifying observable, independent data source, expected observation timeframe and competing explanation; output evaluation plan and unresolved evidence.
Planned.
Read the proposed methodReplicate This Analysis
reproducible input bundle, method/source versions, checksums and CSV/JSON export, with browser-local repeat; output calculation equality and observed source differences.
Shared calculator exercise.
Try Human Revision and Decision RecordPrepare a reviewable policy brief
State the question, affected civilian service, geography and period. Separate dated observations from assumptions. Identify the current baseline and a structured non-AI alternative. Record resources, lawful authority, burdens, fallback and the evidence that would change the decision.
Report unknown costs as unknown. Preserve conflicting claims with their original sources. Obtain the relevant human review before using a calculation to support a public decision. Campaign and candidate research is limited to public policy analysis, not voter profiling or personalized persuasion.
Prepare an executive briefExamine the original MAR assumptionsRelevant calculation engines
Declared fiscal lifecycle ledger
What is the undiscounted and discounted fiscal footprint of the supplied cost ledger?
Unit: minor currency units. Validation: computational unit tests; no empirical calibration.
Open calculator & methodBudget envelope and restricted balances
Which declared accounts have remaining funds or a funding gap after commitments and proposals?
Unit: minor currency units. Validation: computational unit tests; no empirical calibration.
Open calculator & methodEvidence provenance graph
How many declared traceable origin clusters support these reports, and which origins remain unknown?
Unit: records. Validation: computational unit tests; no empirical calibration.
Open calculator & methodSource freshness and comparability
Which records exceed the declared age threshold, have unknown observation dates, or use incomparable periods and units?
Unit: records. Validation: computational unit tests; no empirical calibration.
Open calculator & methodPareto trade-off comparison
Which complete alternatives are no worse on every declared dimension and strictly better on at least one?
Unit: separate declared dimensions. Validation: computational unit tests; no empirical calibration.
Open calculator & methodHuman Revision and Decision Record
What changed, who recorded the decision, why, and which provenance or review records are missing?
Unit: records. Validation: computational unit tests; no empirical calibration.
Open calculator & methodSources and review practice
- NIST: AI Risk Management Framework — Voluntary governance background; not validation of MAR. Reviewed 2026-11-09. Reference, not endorsement or validation.
- U.S. GAO: Cost Estimating and Assessment Guide — Published 12 March 2020. Estimation and sensitivity guidance; no endorsement of this implementation. Reviewed 2026-11-09. Reference, not endorsement or validation.
- UNDRR: Sendai Framework monitoring — Country-reported indicators have coverage limits; no causal program effect is inferred. Reviewed 2026-11-09. Reference, not endorsement or validation.
Does this pathway recommend a political party or candidate?
No. The same inputs and methods produce the same results for every user. No endorsement, voter profile or election prediction is generated.
What if the evidence is incomplete?
Keep missing values unknown, disclose the gap and use the result only within its stated scope. A planned module is not an operational prediction.