0 errors in 64 independently re-verified claims. See how we measure →

Verified, or flagged: never guessed.

Every figure either proves out against a primary source, or comes back with the exact item to pull. Checking the work stops being the work.

It can't invent a number.

It doesn't write the figure. It pulls it, and carries the document, the form and the period it came from — as a live link to the regulator's own site.

Revenue$83,172.5M
FY2025 · 10-K aci-20260228.htm

What the chip carries

Document
aci-20260228.htm
Filer
Albertsons Companies, Inc.
Form · period
10-K · FY2025
Host
www.sec.gov
open the filing on SEC EDGAR

It commits only when its passes agree.

It reads the value more than once and writes nothing when the passes disagree. A slot can never hold two answers, so a disagreement reaches you flagged instead of papered over.

Pass 1 · extract$83,172.5M
Pass 2 · re-check$83,172.5M
Committed · verified$83,172.5M

The passes agree, so the value is written once — with its source attached.

Pass 1 · extract$217.4M
Pass 2 · re-check$2,174.0M
Flagged — needs reviewnothing committed

The passes disagree, so no value is written — the row is handed to a person instead.

One slot, one writer. Only the agreement step can write to it, so nothing overwrites a committed figure later.

It fails closed, and labels the rest.

Every line is proven, quoted, flagged, or absent — and it names the exact item to pull next. Nothing is silently estimated.

Worked exampleAlbertsons · FY2025 · 10-K
Revenue$83,172.5M
FY2025 · 10-K aci-20260228.htm
Net income$217.4M
FY2025 · 10-K aci-20260228.htm
EBITDA$2,641.3M
FY2025 · 10-K aci-20260228.htmoperating income + D&A
Shares outstanding
Flagged — needs reviewCouldn't confirm — two share classes under one standard tag→ pull 10-K cover
Income-tax footnoteNot disclosed
→ pull 10-K Item 8
VerifiedFlagged — needs reviewNot disclosed

What we measured.

64 claims we had rendered as verified, re-fetched, re-parsed and recomputed independently — without looking at the original answer.

0silent errors found
64verified claims independently re-checked
≤ 5.7%95% confidence interval, upper bound

Measured silent-error rate: 0 errors in 64 verified claims, 95% confidence interval upper bound 5.7% — on value, period, and computation, checked against primary sources.

What the screening matcher finds.

A missed designation looks exactly like a clean result, so we measured it against the lists' own alternate spellings. Every screen states these figures in its own limitations, including the class where they are worst.

A spelling the list itself carries24,197 of 24,197 published aliases
100%
A spelling the list does not carry6,556 of 24,187 — the figure that decides a missed designation
27%
A name with the middle name dropped8,220 of 15,260
54%
A name written with initials92 of 15,260 — the worst class, and we say so
0.6%

Measured 2026-07-25 against 20,746 designations on the OFAC and UN lists. Ground truth is the government record: each list publishes its own alternate spellings, so holding one out and asking whether the designation is still reached needs no human labelling. Reproduce it with scripts/measure_matcher.py.

See it run on a company you pick.

Bring a name. We'll run the brief live and open every source with you.

Book a demo