How the platform
knows what it knows.
Definitions, evidence classes, source boundaries and interpretation rules for responsible use of the platform.
Ambition must never
outrun evidence.
Every consequential statement should carry an evidence meaning.
The taxonomy prevents corporate propositions, controlled programme baselines and technical targets from being misread as independently demonstrated performance.
Verified fact
Supported by primary evidence or independent confirmation appropriate to the claim.
Decision-grade where currentControlled baseline
An approved programme assumption with a named owner, status and revision.
Bounded design usePerformance target
An intended technical, financial or delivery outcome that has not been demonstrated.
Validation requiredStrategic proposition
A stated institutional position, ambition, differentiation or future-state thesis.
Not external proofConcept
An illustrative option under assessment, before complete substantiation or approval.
Not a commitmentWhat was
reviewed.
The working corpus contains 121 files organised across eight programme areas.
Sources include 52 DOCX files, 60 PDFs, eight presentations and one engineering SVG. Programme areas include TFOR, TMOR, HTGR SMR, gas-to-nuclear, fusion, data centres, military platforms and supporting institutional materials. Duplicate format editions may represent the same underlying work.
| Rule | Application |
|---|---|
| Currency | Use document revision, issue date and evidence cut-off where stated. |
| Attribution | Separate official corporate claims from independent public records. |
| Geography | Country material does not itself prove approval, contract or deployment. |
| Technical maturity | Use the maturity language in the source; do not infer operating performance. |
| Financial values | Treat projections and capital requirements as forward-looking until verified. |