Reversals

A record kept in good repair. This page is the repair log.

19 entries since July 2026: 7 corrections, 5 status changed after repairs, 2 publication helds, 2 regenerateds, 1 audit, 2 policys. Newest first.

Nothing here is quietly edited. A release that corrects, reverses, holds, or regenerates anything adds an entry before it ships.

If you think a reading of a specific document is wrong, dispute it on that document — every reviewed document carries the link. Disputes are published once reviewed, and when one changes a reading, the change appears here with your objection attached.

The record of our own changes#

This page lists every time the site corrected, reversed, held, or regenerated something it had published — with the date, the reason, and the record where each change is documented. It is updated with every release. It exists because what this site claims is licensed by conduct: the record changes when it is wrong.

Kinds: Correction — something published was wrong and was changed; Status changed after repair — a weekly status moved because the record beneath it changed; Publication held — the site declined to publish until it could count; Regenerated — model-written prose was replaced; Audit — the instrument was tested and the result published; Policy — a decision about how the record is kept.

Ledger#

Date	Kind	What changed	Scale	Why	Evidence
Aug 31, 2026	Policy Research answers — retrieval index quality gate	The rebuilt salience index (junk case citations and extraction artifacts removed) was accepted with two aspirational pool-hygiene thresholds still unmet: 81 aliases shared across four or more questions (limit 10) and 74 documents recurring in three or more pools (limit 25).	—	Both conditions predate the change — the previous accepted index measured 76 and 86 — and the metrics this change could have moved both improved (top-ten arm share 20%→18%, mean relevance 0.49→0.50). The cross-question tail the unmet limits measure is separately scheduled work; the owner accepted the gate under its intent and this entry records the choice.	comment 1 v1.21.2
Aug 31, 2026	Status changed after repair Rulemaking — week of July 6, 2026	The week’s status changed from consistent-with-norms to notable departure, and its narrative was regenerated, after a newly ingested proclamation was read as a clear departure. The executive-actions narrative for the week of July 13 was regenerated for the same ingest (two monument proclamations); its status did not change.	1	Sixteen newly approved subject mappings let the Compilation of Presidential Documents route twenty previously skipped documents — emergency-power renewals, monument modifications, deregulatory orders — into the record; one of them moved a published week’s status. The repair ran under the flip gate and full validation.	798 777 v1.20.3
Aug 31, 2026	Correction Weekly aggregates — in-progress week	Fourteen aggregate rows created early for the in-progress week (August 31) were deleted; the weekly snapshot rebuilds them from complete data.	14	A repair scope mistakenly included the current week, giving it premature status rows — the same artifact corrected on August 28.	777 v1.20.3
Aug 31, 2026	Publication held Weekly digest — week of 2026-08-24	The weekly digest email was held past its Monday send and released the same day, after repair.	—	The snapshot process was killed out of memory before the digest gate ran — and the gate would have held regardless: a late-archived 2024 GAO report had been scored into a baseline week whose aggregate had not been recomputed, one derivation-graph violation. The week was re-derived under the flip gate (zero status changes), validation came back clean, and the digest went out.	777 v1.20.1
Aug 29, 2026	Policy Research answers — release gate	The v1.18 retrieval release was accepted with two eval questions one item lower than the baseline (FW2, IM4).	—	Both lost items were mentions of documents the retrieval had never returned in any run — the reordered pool stopped prompting them. No retrieved document was lost; the owner accepted the gate under its intent and the ledger records the choice.	comment 1 comment 2 v1.18.1
Aug 29, 2026	Regenerated Weekly narratives, January–August 2026	21 category narratives and 21 weekly summaries were regenerated; 2,030 older narratives were accepted as-is rather than regenerated.	42	The presidential-document backfill changed the status of 21 category-weeks, leaving narratives that said "no significant anomalies" beside weeks now at a departure status, and summaries whose category counts no longer matched the data.	comment 1 comment 2 v1.17.10
Aug 29, 2026	Status changed after repair Weekly statuses, January–June 2026	Twenty-one category-week statuses moved up after the backfilled presidential documents were re-derived: elections (01-12, 02-02, 02-16, 03-09, 03-16, 03-30, 06-01, 06-08), civilService (01-19, 06-01), rulemaking (01-19, 02-09, 06-01), judicialIndependence (02-16, 03-30), lawEnforcement (03-30, 06-01), executiveActions (01-05), fiscal (02-09), military (03-30), mediaFreedom (04-13). All accepted.	21	Those weeks had held no presidential documents at all while the ingest was stalled; the documents that arrived — Executive Orders 14399 and 14410 among them — were classified as departures.	comment 1 comment 2 v1.17.8
Aug 28, 2026	Correction Ingest — Compilation of Presidential Documents	Presidential documents issued between 2026-01-07 and 2026-08-28 had never been ingested; 1,018 category rows were backfilled, embedded, and reviewed.	1,018	The publisher loads these documents about seven weeks after their issue date; the weekly step asked only for the previous week, found nothing, and logged that as success for seven months.	798 comment 2 v1.17.8
Aug 28, 2026	Audit AI reviewer — political symmetry	Two hundred current-term verdicts were re-run with the administrations’ names exchanged inside the documents: 11.6% crossed the departure line net of the model’s own noise (19 toward less concerning, 3 toward more). A mirror arm of 190 Biden-era documents renamed to the current administration moved 4.2%.	200	Verdict rates differ sharply by era; the audit tests whether that is the record or the reviewer. Reading: fragility of borderline verdicts to any renaming, not a party tilt — published on the methodology page.	comment 1 comment 2 comment 3 v1.17.3
Aug 28, 2026	Correction AI review annotations	Thirty-three review annotations (18 current-term, 15 baseline-era) were corrected where the reviewer’s summary misstated what the document said.	33	A 400-row sampled audit screened the annotations against their documents; 33 of 100 flagged rows were confirmed substantive errors and rewritten from the document text, with the owner’s approval for the baseline rows.	comment 1 comment 2 v1.17.1
Aug 28, 2026	Correction Court opinions — revisions	Five hundred ninety-two court-opinion rows that duplicated a revised version of the same opinion were marked superseded (never deleted) and dropped from counts; two weekly statuses moved down as a result (immigrationEnforcement 2026-04-06, lawEnforcement 2026-05-04), and seven moved up when emergency-docket orders were added.	592	CourtListener issues a second record when a slip opinion is revised; both had been counted. Emergency-docket orders had been excluded by the search default.	comment 1 comment 2 v1.17.2
Aug 28, 2026	Correction Overview — current week	Fourteen aggregate rows for the week then in progress were deleted after a repair had created them.	14	The repair was run through the current date rather than the last completed week, so the overview showed a partial week as the latest.	DECISIONS.md v1.17.3
Aug 28, 2026	Regenerated Weekly summaries 2026-07-20 and 2026-07-27	Two stored weekly summaries were regenerated.	2	Each named a number of categories and then listed a different number (“Four categories” followed by six). A deterministic number check now runs on every summary before publication.	comment 1 comment 2 v1.17.1
Aug 21, 2026	Status changed after repair Baseline period — GAO reports	One baseline-era weekly status changed when GAO reports were added back to 2017.	1	A new source is required to run uniformly across every period; the owner reviewed and accepted the single flip.	739 DECISIONS-ARCHIVE.md v1.15.1
Aug 11, 2026	Correction AI review annotations	Five review annotations were corrected, beginning with one that said Executive Order 14029 “revokes Schedule F”.	5	Answer audits found annotations that misread their documents; the corrections were applied from the document text and the sampled audit at scale was scheduled.	comment 1 comment 2
Aug 10, 2026	Publication held Weekly digest	The weekly digest was held and released by hand after the missing scores were computed (also on 2026-08-03).	—	Documents that their sources published late were stored outside the two-week scoring sweep and nothing scored them; the integrity check refused to send a digest built on unscored documents. The snapshot now reconciles unscored documents itself.	667 DECISIONS.md
Aug 2, 2026	Status changed after repair Baseline periods — inspector-general reports (oversight.gov)	Thirty-one baseline-era weekly statuses changed when reports from seven inspectors general were added across every period.	31	Uniform coverage across periods is required of every new source; the owner reviewed and accepted the flips.	655 DECISIONS-ARCHIVE.md v1.5.0
Jul 30, 2026	Status changed after repair Baseline periods — congressional hearing transcripts	One hundred one baseline-era weekly statuses changed when hearing transcripts were added across every period.	101	Uniform coverage across periods; the owner reviewed the flips (for example, the 2018 family-separation week) and accepted them as mission-correct.	DECISIONS-ARCHIVE.md
Jul 15, 2026	Correction Press Freedom category	Sixty-seven documents were dropped from the Press Freedom category by a relevance filter; a fresh holdout week showed no false drops.	67	Unrelated documents had been routed into the category and were contaminating its counts; the filter was adjudicated document by document (50 of 50 labels agreed) before it shipped.	543 544 DECISIONS-ARCHIVE.md