GPS-bench/gps-bench-ai-bills

GPS-Bench: AI Bills and Expert Analyses

984 AI-governance instruments with their full text, and 398 independent expert analyses covering 241 of them, organised for leakage-controlled evaluation.

The unit of this dataset is not a bill. It is a bill paired with what human experts said about it. A model can be asked what an instrument requires, who it covers, or what it will do — and scored against analyses written by lawyers, think tanks and civil-society researchers who had no idea a model would be graded on their words.

What is here

File	Rows	Contents
gps_us_ai_bills.json	838	US federal, state and city measures
gps_international_ai_bills.json	98	National laws and frameworks outside the US, EU and China, plus multilateral instruments
gps_china_ai_bills.json	44	PRC laws, administrative regulations, ministry notices and GB standards
gps_eu_ai_bills.json	4	EU regulations, directives and codes of practice
gps_bench_bills_WITH_analyses.json	209	The analysed subset, grouped by leakage split — the evaluation set

The four bloc files partition the corpus: every bill appears in exactly one, with no duplicate bill_key within or across them. gps_bench_bills_WITH_analyses.json is a view, not a fifth partition — it re-groups the analysed bills by split.

That view lags the slices. The bloc files hold 241 bills with at least one analysis; the published view holds 209. It is a snapshot from before the 2026-08-10 collection round, kept deliberately so existing code and any published results keep resolving to the set they were computed on. If you want the current analysed subset, derive it from the slices with n_human_analyses >= 1 rather than reading the view.

Total bill text: 46,363,560 characters. Every row carries full text: the last metadata-only row, North Carolina H301, was removed rather than kept without the statute. All 984 rows carry an exact YYYY-MM-DD date, and bill_date_basis says which milestone each one marks — an introduction date and a promulgation date are not the same measurement, so read it before comparing dates across jurisdictions.

Two date caveats, both left visible on the row rather than smoothed away. Where an AI provision was inserted into a much older statute, bill_date carries the amendment and host_statute_date keeps the original: Singapore's Parliamentary Elections Act is dated 2024-10-15 for its deepfake amendment, not 1954. Two Japanese rows have the same shape but no exact amendment date in any source, so they keep their host-statute date (1970, 2003) and carry ai_provision_year instead — filter on ai_provision_added_later to find all three. And two Georgia rows carry a future commencement date with bill_date_basis: effective; their signing dates were not collected.

US coverage is all 50 states plus the District of Columbia (214 state rows, 623 federal, 1 city). Wisconsin and Wyoming were the last two gaps and are now filled: 2023 Wisconsin Act 123 and 2025 Wisconsin Act 34, and Wyoming's 2026 Chapter 91 alongside HB 91, which failed in committee and is kept as a negative case. A state row is kept when an AI term appears in the bill's title; matching the body instead returns appropriations acts. That rule is applied consistently across the states but not across the federal subset, which was collected earlier under a looser rule — see Known gaps.

Instruments per country

41 countries, plus the EU and one multilateral instrument reported separately — no member state's count absorbs the AI Act. "Live" excludes measures that failed or were vetoed (is_live); "benchmark" is live rows carrying two or more independent analyses. Regenerate with python3 count_by_country.py.

Country	Instruments	Live	Benchmark
United States	838	610	76
China	44	44	16
United Kingdom	18	8	5
Canada	12	11	3
India	8	8	7
Singapore	8	8	6
Japan	4	4	3
Kenya	4	4	0
Saudi Arabia	4	4	4
European Union	4	3	4
Thailand	3	3	1
France, Ireland, Nigeria, Russia, South Korea	2 each	2 each	0–1 each
Australia, Brazil, Denmark, Germany, Italy, New Zealand, Qatar, Spain, Taiwan, UAE, Vietnam	1 each	1 each	1 each
Argentina, Chile, Colombia, Egypt, Indonesia, Iran, Israel, Malaysia, Mexico, Netherlands, Pakistan, Peru, Rwanda, South Africa, Turkey	1 each	1 each	0
Multilateral	1	1	1
Total	984	744	139

The 240 excluded are 228 US (220 federal bills that died when their Congress adjourned, 2 vetoed, 6 state — including Wyoming HB 91), 10 UK measures that fell at session end, the withdrawn EU AI Liability Directive, and Canada's AIDA (C-27), which died with the 44th Parliament.

The 15 countries in the last row are newly collected and hold bill text without commentary yet. They are real coverage for corpus-level work and not usable as benchmark items, because a row with no analysis has no answer key.

Loading

The bloc files wrap their rows in a metadata object, so read the array by key rather than expecting a bare list:

import json, urllib.request

URL = "https://huggingface.co/datasets/GPS-bench/gps-bench-ai-bills/resolve/main/"

us = json.load(urllib.request.urlopen(URL + "gps_us_ai_bills.json"))["bills"]
cn = json.load(urllib.request.urlopen(URL + "gps_china_ai_bills.json"))["instruments"]

# The evaluation set is keyed by split, not wrapped in one array.
ev = json.load(urllib.request.urlopen(URL + "gps_bench_bills_WITH_analyses.json"))
after = ev["after_cutoff"]

# Filter on status_class, never on bill_status -- the latter keeps each collector's own
# words, so "is this law?" needs the normalised bucket. This works identically across
# all four blocs; matching bill_status strings does not.
enacted_us = [r for r in us if r["status_class"] == "enacted"]        # 163
enacted_cn = [r for r in cn if r["status_class"] == "enacted"]        # 29

# "Exclude failed and vetoed" is `is_live`, NOT status_class. A federal bill that died
# when its Congress adjourned still carries status_class "active", because that field
# records the furthest procedural stage reached, not the final outcome. Filtering on it
# counts 220 dead federal bills as live.
live = [r for r in us + cn if r["is_live"]]

gps_china_ai_bills.json uses instruments, not bills, because Chinese AI governance is issued as regulations, notices and standards rather than as bills passed by a legislature. The field schema is otherwise identical.

The evaluation set

Across the bloc slices, 241 bills carry at least one expert analysis and 139 carry two or more. The published evaluation-set file is a snapshot of that set taken before the 2026-08-10 round, so it holds 209 and 128 respectively. Both numbers are real; which one applies depends on whether you read the slices or the view.

gps_bench_bills_WITH_analyses.json groups its 209 by split:

Bucket	Bills	Meaning
before_cutoff	57	Earliest analysis predates the reference model's training cutoff. The memorisation baseline — a model may have read the commentary.
after_cutoff	50	Earliest analysis postdates the cutoff. Leakage-free for the reference model; the generalisation set.
train	20	Deterministic hold-out for few-shot prompting or calibration. Never score on this.
one_analysis	82	Held apart: a single analysis is real evidence but not a two-annotator item.

The 32 bills added since — Gulf, Latin American and other newly collected rows, plus US and China bills that gained a first or second analysis — are in the slices with their analyses and splits, but not yet in this file. Rebuild it with python3 export_merged_analyses.py when you want them in the evaluation set.

Every benchmark-eligible row carries a split. This was once a live gap — 47 rows had none, 43 having gained a second analysis after assign_splits() last ran, and 4 dropped by a builder that tested n_human_analyses != 2, an exact inequality that ejected any bill attracting a third analysis. That test now reads >= 2, and splits were assigned without re-deriving the hash hold-out, so no already-published row changed condition.

Splits are computed against claude-opus-4-5-20251101, training cutoff 2025-05-31, recorded per row in split_reference_model / split_reference_cutoff. The split is only meaningful relative to that model. For any other model, use the per-row leakage_free_for list, which names the models whose training cutoff strictly precedes the bill's earliest analysis. Models with an unpublished cutoff are omitted from that list rather than guessed at — an unknown cutoff cannot support a leakage-free claim.

Analyses per bill across the slices: 121 bills with two, 18 with three, 102 with one. A third analysis never excludes a bill from the benchmark.

Row schema

Every row carries the same core fields, in the same order, and may add optional per-row provenance after them (DOIs, archived URLs, translation blocks, recovery notes). The contract is machine-checkable: schema/gps_slice.schema.json states the required fields and types, and validate_slices.py checks every slice against it plus the invariants a schema cannot express — bill_key uniqueness, n_human_analyses agreeing with the slots beneath it, and earliest_analysis_date being the earliest date actually present. Absent values are the empty string, never null; the two exceptions are documented in the schema.

The bill

Field	Meaning
bill_key	Stable slug, e.g. co-sb205, cn-genai-interim-2023
bill_name	Official title
jurisdiction	federal, state:XX, city:XX, eu, national:XX, international
bill_date	Exact YYYY-MM-DD on all 984 rows. Which milestone it marks varies by row and is stated in bill_date_basis, never left to be guessed.
bill_date_basis	What bill_date actually is: introduced, first_action, first_reading, first_consideration, executive_signature, chapter_assigned, promulgated, translation_published, Royal Assent and a few source-specific phrasings. Read this before comparing dates across jurisdictions — an introduction date and a promulgation date are not the same measurement.
bill_date_source	Which record the date came from, down to the cache file.
enacted_date / effective_date	Present where the milestone differs from bill_date and is known. Added rather than overwriting, so a row can carry both when it was introduced and when it became law.
prefiled_date / filed_date	Present on the handful of rows where the source distinguishes them from first reading.
bill_status	Procedural status exactly as its collector wrote it — 71 distinct values across the four files, and deliberately not normalised. Filter on status_class instead.
status_class	The one regime-independent bucket: enacted, active, stalled, guidance, draft, other. Every count this project publishes, including the enacted totals, comes from this field, so it is shipped rather than left to be re-derived. See below for what it does and does not claim.
bill_text	Full text of the instrument
bill_text_source / bill_text_status	Where the text came from and whether it was fetched
bill_language	Measured from the characters, not inferred from the jurisdiction
bill	Composed header + full text, for single-string prompting
group	CAST catastrophe family: misuse, loss_of_control, systemic, geopolitical
group_basis	author_coded_heuristic — a hypothesis for the analyses to adjudicate, not a label
phase	1_adoption, 2_implementation, 3_societal

Progress, sponsor and outcome — added so measures can be compared across legislatures rather than only read one at a time

Field	Meaning
status_text	Where the measure got to, on one ladder: Introduced or Prefiled → Out of Committee → Passed Original Chamber → Passed Second Chamber → Sent to Executive → Enacted → Effective, plus the terminals Vetoed and Failed. This does not replace bill_status, which keeps whatever its collector said — three vocabularies, including raw govinfo version codes.
status_rank	status_text as a sortable integer. 1–7 walk the ladder, -1 is a terminal negative, 0 is unknown. The sign is load-bearing: ordering by it never claims a vetoed bill got further than an enacted one.
status_text_source	Which derivation produced it, with evidence — a tracker Status Text, a classified OpenStates action, a congress.gov latestAction, or the coded bill_status.
primary_party	Party of the primary sponsor: Democrat, Republican, Independent, Non-Designated. Non-Designated is a positive finding (nonpartisan chamber, committee bill, tracker declined to code) and is not the same claim as "", which means nobody has looked.
sponsor_party_source / sponsor_party_note	Where the party was read from, and how it was determined.
outcome_reason_kind	enacted, failed, vetoed, or "" for a measure still moving. The categorical companion to the prose in why_enacted, why_failed and outcome_detail, which cannot be filtered on.
is_live	The field to filter on for "exclude failed and vetoed." False when the final outcome was failure or a veto, or the measure is stalled or a spent draft; true otherwise. 744 of 984 rows are live. It exists because neither status_class nor outcome_reason_kind answers this alone — see the caution below.
key_actors	Who sponsored, backed, or is regulated by the measure.

Two tiers, because only one of them can be normalised across regimes. Whether a rule binds is a question every legal system answers, so China's in_force, the UK's Royal Assent and a US enacted all collapse to status_class == "enacted" and are safely comparable. The procedural ladder in status_text is a different matter: Passed Original Chamber → Passed Second Chamber presumes a bicameral legislature, and the NPC Standing Committee, a ministry 通知 and an EU trilogue have no honest position on it. That is why status_text is populated on the US slice and left empty on the other three — an absence by design, not a collection gap. Compare across blocs on status_class; use status_text only within a bloc.

status_class is not an outcome field, and this trips people up. It records the furthest procedural stage a measure reached, so 220 federal bills that died when their Congress adjourned still read active — their last recorded action really was "introduced / referred", and no later action exists to overwrite it. Both fields are correct as defined; they answer different questions. outcome_reason_kind gives the final outcome but is empty or absent on 537 rows, so it cannot carry a filter by itself either. is_live folds the two together and is the only field that answers "is this measure still capable of taking effect, or already law?" across all 984 rows. Counting live instruments by status_class == "active" overstates the US by 220.

Three cautions on status_class. guidance marks instruments that were issued but do not bind — advisory frameworks and ministry guidelines, 5 rows, held out of enacted because they are not law and out of other because they were in fact issued. And 110 federal rows reached us with a govinfo document version code in bill_status (ih (congress 118, hr), rs (congress 118, s)) in place of a status. Each such code names a pre-passage stage, so none can mean enacted and all classify as active — but that is the floor, not a recovered outcome. The status those codes overwrote is gone until the row's text is re-fetched, so treat active on a federal row as "not shown to be law", not as "still moving".

Coverage is uneven, and left visible rather than smoothed over. status_text resolves on 810 of 838 US rows; the 28 gaps are rows whose only recorded status is other or in_progress, and they are empty rather than guessed. primary_party is set on 602 of 605 federal rows but only 66 of 210 state rows, because federal sponsors come from the govinfo BILLSTATUS XML while state sponsors need the OpenStates sponsorships endpoint, which the cached sweeps never requested. key_actors, why_enacted and why_failed exist only on the 397-row curated federal subset.

The analyses — N in 1–3

Field	Meaning
human_analysis_N	Verbatim excerpt. The citable ground truth.
human_analysis_N_full	Full article text, for context
analysis_N_source / _author / _url / _date	Citation
analysis_N_source_type	law_firm_legal_analysis, think_tank, civil_society, news_media, academic_analysis, industry_analysis, other_expert_analysis
analysis_N_lean	Recorded ideological lean, where known — for controlling viewpoint, not for filtering
analysis_N_date_precision	day, month, year or none. 337 analyses carry a full date, 6 resolve only to a month, 7 only to a year and 4 have none — law-firm alerts routinely publish "December 2025" with no day, and four sources are continuously-updated trackers with no publication date at all. The partial date is kept verbatim rather than having a day invented for it. The field is not yet set on 44 of the 398 slots, all added in the 2026-08-10 round; absent precision is not the same as absent date.
analysis_N_date_note	Why a date is partial or absent, where it was checked.
n_human_analyses	How many slots hold a valid analysis
earliest_analysis_date / _precision	Earliest date across all three slots, and how precisely it is known. Drives the leakage split.
leakage_free_for	Model ids that could not have memorised the commentary
leakage_computable	Whether the leakage claim could be computed at all. Read this before trusting an empty leakage_free_for: an empty list means "computed, no model qualifies", which is a different claim from "could not be computed". Three rows are false because an analysis carries no publication date, leaving the earliest commentary unbounded below. Absent on rows with no analyses.

How a partial date is compared to a cutoff. A leakage-free claim requires the model's cutoff to precede the analysis, so a partial date is tested at the earliest instant it could denote — 2025-12 is compared as 2025-12-01. Bounding the analysis as early as possible can only withdraw a claim, never grant one that has not been earned.

Integrity rule, inherited from CAST: no excerpt, no analysis. A row without a source name, a URL and a verbatim excerpt is not counted. Nothing in any analysis field is generated, summarised or paraphrased — every excerpt is quoted from the real article.

Who wrote the analyses

Source type	Count
law_firm_legal_analysis	165
think_tank	72
news_media	44
industry_analysis	23
consulting_vendor	18
academic_analysis	16
other_expert_analysis	15
civil_society	9
civil_society_policy_analysis	7
independent_policy_analysis	7
government_official_interpretation	5
legal_regulatory_explainer	5
Eight further types, 1–3 each	12

The vocabulary has drifted and is not yet collapsed. 20 distinct values are in use against the 7 the collection script offers, and several are one category wearing different names — academic_analysis, academic_legal_analysis, academic_expert_analysis and peer_reviewed_legal_scholarship are all academic work, 21 slots split across four labels. Group them before filtering or reporting, or a stratified sample will silently miss most of a category.

Five analyses are tagged government_official_interpretation — a ministry or regulator reading its own instrument. They are not independent commentary and are labelled so they can be excluded; among them, MIIT on its 2018 AI work plan, the CAC's technology centre on the Clear 2022 algorithm campaign, and Canada's Information Commissioner on Bill C-58, which empowers her own office.

This distribution is itself a limitation. Nearly half the analyses are law-firm client alerts, a genre with a de-facto shared structure — what the law does, who is covered, obligations, effective dates, action items. Two annotators drawn from one genre agree cheaply, which is exactly what a two-annotator design is meant to test. analysis_3 exists to add a recorded, different viewpoint; analysis_N_source_type and analysis_N_lean let you stratify or restrict rather than take agreement at face value.

Languages

964 of 984 rows hold English in bill_text. The remaining 20 are in 13 other languages, and every one of them now carries English beside it — there is no row in this corpus a non-reader of the source language cannot work with:

Source language	Rows
Spanish	6
French, Thai	2 each
Danish, Dutch, Indonesian, Italian, Japanese, Korean, Persian, Portuguese, Russian, Vietnamese	1 each

No row is Chinese-only: the last such row, MOST's 2019 AI pilot-zone Guidelines, was moved onto CSET's translation, with the authoritative Chinese kept beside it in bill_text_zh_primary.

Where a row's text is not English, the original stays authoritative in bill_text and English sits beside it in bill_text_en_machine (machine translation) or bill_text_en_published (a third-party human translation), with method and provenance in a per-row translation object. A machine translation of a statute is not the statute: a mistranslated scope clause or penalty threshold reads perfectly fluently, and a model graded on "what does this provision require" would be scored against the translation rather than the law.

Three rows carry bill_text_zh_primary — the authoritative Chinese original beside a third-party English translation, because PRC laws and regulations carry no copyright under Article 5 of the PRC Copyright Law while the English renderings do.

The machine translations were produced by four different models, because the queue was split across providers to run concurrently: gpt-4.1 (6 rows), claude-sonnet-4-5 (5), deepseek-v4-pro (5), gemini-3.6-flash (3). Each row's translation.model and translation.provider record which produced it. Treat that as a caveat, not a detail: translator identity varies across rows, so an experiment that scores models on the English text is measuring translator variance alongside whatever it means to measure. For anything load-bearing, score against bill_text.

One row is a worked example of why the redundancy matters. Thailand's ETDA report was refused by one provider — two chunks returned finishReason=OTHER — and translated without incident by another. A refusal is a property of the vendor, not of the document.

Curation

Bill text comes from official sources wherever one exists: Congress.gov, OpenStates, EUR-Lex, national gazettes, cac.gov.cn, most.gov.cn. Where an official URL had rotted, the recovery is recorded on the row rather than silently substituted — the China pilot-zones instrument, for example, carries a source_recovery block noting that all twelve URLs in the source product now 404 and which mirror was used instead.

Analyses were collected by hand and by a proposal pipeline that searches a registry of publishers, fetches each hit, and verifies the bill is actually named in the article body before proposing it. A search engine matching "SB 1047" proves nothing until the text is read. The pipeline only ever proposes; a human reads the article and chooses the excerpt.

Article counts, where present, are computed by walking the numbering in document order and counting openings — not by counting regex hits, which overcounts when an article cites other articles, when an instrument cites another instrument's articles, or when a bilingual row states each article twice.

Known gaps

Stated plainly, because they bound what this dataset can support:

• 743 of 984 bills have no analysis at all. 689 of those are US. The corpus is much larger than the evaluation set, and most of it is bill text without commentary.
• The 845 rows with fewer than two analyses have no split, which is by design: the split is an evaluation-set property. Every one of the 139 benchmark-eligible rows now carries one. Filter on n_human_analyses >= 2 for evaluation. (bill_date was in this list until every row was dated; all 984 now carry an exact date, with bill_date_basis saying which milestone each marks.)
• Coverage is skewed to the US (838 of 984) and, within it, to federal bills.
• The inclusion rule is not consistent between the federal and state subsets. A state row needs an AI term in its title, and 198 of 214 satisfy that. Only 360 of 623 federal rows do. The other 263 federal rows would fail the rule applied to the states, and because they are omnibus vehicles — NDAAs, an FAA reauthorisation, a reconciliation act — they carry 22.2M characters, about 48% of all text in the corpus, for provisions where AI is one subject among dozens. Any length-weighted statistic is dominated by them. Filter on a title match, or on ai_relevance, before computing one.
• State selection favoured enacted law. External trackers record roughly 1,200 AI bills introduced across the states in 2025 and about 145 enacted; this corpus holds a small share of the former and most of the latter. It is close to a census of enacted state AI law and a thin, non-random sample of what was merely introduced. Do not read state counts here as a measure of legislative activity.
• Many bill_date values on China rows are translation dates, not issue dates. Rows sourced from CSET carry bill_date_basis: translation_published, and the gap runs to years: the 2022 scenario-innovation Guiding Opinions carry a late-2025 date because that is when the English appeared. split derives from bill_date, so on those rows it describes the translation's timing rather than the instrument's. Check bill_date_basis and original_publication_date before treating bill_date as the date Chinese actors first saw the policy.
• One row (nyc-ll144) has two analyses whose source pages carry no publication date. It counts toward the two-analysis total but contributes nothing to the leakage split, and its leakage_free_for is deliberately empty.
• The (Trial) Measures for AI Technology Ethics Review row runs to Article 37 but opens only 36 — the position where Article 9 should stand reads Article 4 in the source PDF. Recorded, not silently corrected.

Licensing — read before redistributing

This dataset mixes three rights regimes and is not uniformly redistributable.

1. Bill text is generally not copyrightable. US federal and state legislation, EU instruments, and PRC laws and regulations (Article 5, PRC Copyright Law) are public texts.
2. human_analysis_N_full contains 398 whole third-party articles from 277 distinct sources — 4,025,046 characters of law-firm client alerts, think-tank commentary and trade-press writing. These are copyrighted works reproduced in full. They are included for research reproducibility. Redistribution, commercial use, or training on this field may not be covered by fair use or fair dealing in your jurisdiction. The verbatim human_analysis_N excerpts are short quotations and sit on much firmer ground than the full texts.
3. Five China rows hold English pasted from ChinaLawTranslate, whose robots.txt sets Content-Signal: ai-train=no. Their text predates this collection and is retained as published; two now carry the public-domain Chinese original in bill_text_zh_primary, and three do not. Per-row terms are in bill_text_access_note.

If you need a cleanly licensed corpus, drop human_analysis_N_full and keep human_analysis_N plus analysis_N_url — the excerpt and the citation.

Intended use, and what this is not

Built for evaluating LLM reasoning about AI governance: what an instrument requires, who it binds, how it interacts with neighbouring law, and what experts judged its effect would be. The leakage split is the point — scoring only on before_cutoff measures recall of published commentary, not analysis.

This is not legal advice, not a complete census of AI legislation, and not a gold-standard annotation set. The analyses are expert commentary, not adjudicated truth; two experts frequently disagree, and group is an author-coded hypothesis the analyses are meant to test rather than a validated label.

Citation

@misc{gpsbench_ai_bills,
  title  = {GPS-Bench: AI Bills and Expert Analyses},
  author = {Melanie},
  year   = {2026},
  note   = {Dataset of AI-governance instruments paired with independent expert analyses,
            organised for leakage-controlled evaluation},
  url    = {https://huggingface.co/datasets/GPS-bench/gps-bench-ai-bills}
}

Please also cite the original publishers of any analysis you quote — each row carries analysis_N_source, analysis_N_author and analysis_N_url.

Changelog

2026-08-10

• Corpus grew from 952 to 984 instruments, and analyses from 356 to 398 on 241 bills. Almost all of the growth is international: that slice went from 54 to 98 rows and from 14 countries to 41, adding the Gulf (Saudi Arabia, UAE, Qatar, Egypt), Latin America (Chile, Colombia, Peru, Argentina, Mexico), and Indonesia, Iran, Israel, Malaysia, Pakistan, South Africa, Taiwan, Thailand, Turkey.
• Every non-English row now carries English beside it. 20 rows in 13 languages hold bill_text_en_machine or bill_text_en_published; bill_text is untouched and every bill_text_sha256 still validates. The machine translations come from four different models — see Languages for why that is a caveat and not a footnote.
• gps_bench_bills_WITH_analyses.json is unchanged and remains the evaluation-set file. It now lags the slices by 32 bills and 44 analyses, because this round's collection is not yet exported into it. That is deliberate: existing code and any results already computed keep resolving to the same set. Rebuild with export_merged_analyses.py when you want the new bills evaluated.
• Dating is no longer a known gap: all 984 rows carry an exact YYYY-MM-DD date.
• New caveat recorded rather than fixed: analysis_N_source_type has drifted to 20 distinct values against the 7 the collection script offers, with academic work split across four labels. Group before stratifying.

2026-08-08

• Corpus grew from 922 to 952 records. Eleven came from the international ingest — Kenya (4), Nigeria (2), Rwanda (1) and four UK measures — and nineteen were recovered from the local benchmark build, which had rows present in no published slice. The recovered rows are tagged merged_from and carry fewer fields than their neighbours.
• Added the normative layer to every slice: normative_character (hard_law / soft_law / non_instrument) and the coverage_area block with its classifier basis, confidence and runner-up. coverage_area is the sector an instrument governs, never its geography.
• Ten UK measures are now is_live: false with outcome_reason_kind: failed_session_end — they fell when their legislative session ended rather than being defeated on the merits.
• All four slices now share one key order and pass validate_slices.py. Ten fields the card already documented were missing from schema/gps_slice.schema.json, so nothing validated them; they are now declared, which is what caught a boolean in an object field and an out-of-enum status value.
• Known gap: the nineteen recovered rows have no coverage_area yet. It comes from a semantic classifier and was left empty rather than guessed.