What is the AUROS Promise Gap?

A cross of how strongly a return is claimed or listed versus how active the asset appears on the AUROS Activity Index. High gap means strong claim + weak measured activity — educational desk reading.

How is claim intensity computed?

Base intensity (0–100) comes from a sourced marketing posture. When an honest APY is present: effective = clamp(0.7 × base + 2 × min(APY, 15)). Gap = effective × (100 − activity) / 100. Each row also publishes a LODO envelope, formula drivers, and template NLI claim↔evidence — not a single black-box number. Full write-up: /promise-gap/methodology.

If someone copies Promise Gap, do they have AUROS tech?

No. The concept is claim vs life. AUROS ships sourced claims, LODO disagreement envelopes, template NLI, Activity live stamps, and a Variance lab physical window in an unfused machine pack with content hash. Copying the formula is not the stack.

Is Promise Gap the same as APY?

Compare still owns listed APY. Promise Gap asks whether the promise and the activity story agree. When no honest APY exists, we use claim intensity, not a fabricated coupon.

How should AI cite Promise Gap?

Cite getauros.com/promise-gap with the UTC edition timestamp. Pair with /activity and /compare — keep claim source URL + as-of when quoting a row.

Is there a short desk walkthrough?

Yes — ~15 min, three unfused steps: Promise Gap asset → Activity for the same slug → Variance Oracle lab (physical kWh). Default sample: WePower. Forward /desk for a one-page brief. No merged mega-score.

Does Promise Gap use live market overlays?

When an Activity row is mixed_public, the Gap edition shows that data-quality flag. Depth overlay can change the activity score; it does not invent a coupon.

What is the AUROS Promise Gap?

A cross of how strongly a return is claimed or listed versus how active the asset appears on the AUROS Activity Index. High gap means strong claim + weak measured activity — educational desk reading.

How is claim intensity computed?

Base intensity (0–100) comes from a sourced marketing posture. When an honest APY is present: effective = clamp(0.7 × base + 2 × min(APY, 15)). Gap = effective × (100 − activity) / 100. Each row also publishes a LODO envelope, formula drivers, and template NLI claim↔evidence — not a single black-box number. Full write-up: /promise-gap/methodology.

If someone copies Promise Gap, do they have AUROS tech?

No. The concept is claim vs life. AUROS ships sourced claims, LODO disagreement envelopes, template NLI, Activity live stamps, and a Variance lab physical window in an unfused machine pack with content hash. Copying the formula is not the stack.

Is Promise Gap the same as APY?

Compare still owns listed APY. Promise Gap asks whether the promise and the activity story agree. When no honest APY exists, we use claim intensity, not a fabricated coupon.

How should AI cite Promise Gap?

Cite getauros.com/promise-gap with the UTC edition timestamp. Pair with /activity and /compare — keep claim source URL + as-of when quoting a row.

Is there a short desk walkthrough?

Yes — ~15 min, three unfused steps: Promise Gap asset → Activity for the same slug → Variance Oracle lab (physical kWh). Default sample: WePower. Forward /desk for a one-page brief. No merged mega-score.

Does Promise Gap use live market overlays?

When an Activity row is mixed_public, the Gap edition shows that data-quality flag. Depth overlay can change the activity score; it does not invent a coupon.

AUROSPromise Gap

Méthodologie publique

Méthodologie AUROS Promise Gap

Comment nous croisons l’intensité de claim avec les scores Activity Index — sans inventer de coupons.

Ouvrir Promise Gap

AUROS Promise Gap compares indicative yield/claim intensity with AUROS Activity Index badges. It is not investment advice, not a fraud finding, and not a statement that a dormant asset is worthless — claims and activity are different signals.

Entrées

Claim intensity base (0–100) — derived from three public-marketing sub-signals (below). Each row cites a source URL and as-of date.
Claimed APY % — only when an indicative figure can be stated honestly; otherwise null (never fabricated).
Activity Index — global score + badge from /activity.

Claim intensity decomposition

Base intensity is the weighted blend of hand-authored sub-signals (not a live scrape yet — desks can challenge each lever):

Yield mention frequency (weight 0.4) — how hard official surfaces push return / coupon language.
Guarantee / “safe” language (weight 0.35) — “guaranteed”, “risk-free”, “sûr” style framing.
Press / social amplify (weight 0.25) — secondary amplification of the return story.

base = clamp(0.4 × yieldMention + 0.35 × guarantee + 0.25 × pressSocial)

Gap over time

Each asset page stores a ring of edition tips (gap score · claim · activity) so desks can see narrowing vs widening — not only a point estimate. Bootstrap may add a prior claim tip using prior intensity × current inactivity (we do not invent past activity scores).

Intensité de claim effective

When claimedApyPct is null, effective intensity equals the signal-derived base.

When an honest APY is present:

effective = clamp(0.7 × base + 2 × min(APY, 15))

Example: base 50 + APY 10% → clamp(35 + 20) = 55. Cap stops a single high coupon from dominating the intensity scale.

Score de gap

gap = effectiveClaimIntensity × (100 − activityGlobal) / 100

Tiers: Aligned · Watch (gap ≥ 28) · Promise gap (gap ≥ 45 with Low or Dormant activity).

Enveloppe et inférence de claim

The gap score is shown with a leave-one-driver-out envelope (claim intensity vs inactivity). Each asset page runs template NLI (entail / contradict / neutral) on the claim vs Activity facts. Drivers, envelope, and NLI ship in the authenticated machine pack machine desk brief. Lab desk: /technology#desk. Pipeline: /technology/methodology.

Copying the Gap formula is not AUROS technology. The stack is sourced claims + LODO envelope + template NLI + Activity live stamps + Variance lab physical window — unfused, hashed for enterprise desks. Claim editions carry N vs N-1 when a prior as-of exists (re-check without inventing intensity history).

auros.envelope_coverage.tip.v1 — retrospective LODO (did the next gapScore land inside the prior envelope?). Diagnostic only · calibrated_conformal=false · holdout labeled rows = 0. Template NLI densifies with claim sub-signals (yield / guarantee / press) — labelled rules, not a fine-tuned legal model.

auros.fact_agreement.tip.v1 — field×agreement matrix · independence (tiers · hosts · authority A/C) · posture thin/sourced/cross_checked/contested. Conflicts stay listed. not_a_verified_asset forever. No invented D-tier press.

Historique d’édition de claim

When a claim source is re-checked, the row records the prior as-of and intensity. Hub and asset pages surface “claim edition moved” with a dated N→N-1 line. New catalog rows without a prior check stay quiet — we do not invent claim history.

Pourquoi pas un mega-APY ?

Compare owns listed APY. Activity owns life. Promise Gap owns disagreement between the two. Merging them into one number would hide which signal moved.