{ "row": "data-provenance-explorer", "entity": "data-provenance-explorer", "address": "data provenance explorer", "addresses": [ { "address": "data provenance explorer", "slug": "data-provenance-explorer-aio-20261029", "addr_id": "ADDR-6e5a719fee75", "obs_id": "OBS-dc0adc018f10", "axn": "AXN:0AAC.CAPTURE.🕔🎨🟢●🌺☉" } ], "epoch": "2026-10-29", "type": "C (public entity on which the archive bears: a database and its explorer tool; an entity of the MMRS meta-aggregator)", "surface": "Google AI Overview (expanded)", "auth": "signed out, incognito", "status": { "procedure": "EA-NEGONT-02 v0.9, #1690: equal case, ρ = extends, read from the MMRS record, the capture record and #725's statement of relation; archive selected by D/R/O (10 deposits admitted by reading, 6 with claims and 4 grouped as instances; hop of 73 read and rejected), with the MMRS record (21 claims) and the Capture Registry's findings and analysis (9 claims) ledgered as the archive's records of the entity; the field B by three origins (6 cards, 4 own pages, 2 Wikipedia titles); the composer read the ledgers, never the transcript, except where the brief asks for T's claims and the delta", "ledger": "unaudited (§3.8: one extractor per ledger); field quotes 48/48 verified by script against saved text, the bundle's extracted page text, the data file's header, the Wikipedia responses or the card text; archive quotes 49/49 verified against the registry, census.json, articles.py and the deposit texts; MMRS-17 and MMRS-13 carried with defects noted", "frozen": false, "breaches": [] }, "spec": "EA-NEGONT-02 v0.9, #1690", "objects": { "T": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/T1-data-provenance-explorer-aio-20261029.txt", "sha256": "314d4267ef914693374074841f3ff4e1547a4f6dd73951a6c2b22bcdc5043307", "words": 110, "cards": 6, "words_note": "one transcript at one address (110 words in the answer body, links reduced to their text, markers and the card rail excluded; 6 cards)", "claims": [ "T1-1 definition: 'an interactive audit tool built by the Data Provenance Initiative to track, inspect, and filter large language model (LLM) training datasets' (three markers, one of them the Initiative's link)", "T1-2 dataset explorer: 'Dataset Explorer: Search and filter nearly 4,000 text, speech, and video datasets by creators, original data sources, geographies, modalities, and derivation chains.'", "T1-3 licence filtering: 'License & Use Case Filtering: Evaluate data strictness ranging from commercial use to academic-only or non-commercial restrictions.' (one marker)", "T1-4 web data: 'Web Data & Consent Explorers: Track shifting data-use preferences and consent signals across thousands of web domains commonly scraped for AI training.'", "T1-5 closer: 'Would you like to search for a specific dataset or learn how to filter for commercial-use licenses in the explorer?'" ], "register": { "register": "data/EA-WG-CAPTURES-01.json", "slug": "data-provenance-explorer-aio-20261029", "obs_id": "OBS-dc0adc018f10" }, "transcripts": [ { "slug": "data-provenance-explorer-aio-20261029", "address": "data provenance explorer", "addr_id": "ADDR-6e5a719fee75", "obs_id": "OBS-dc0adc018f10", "axn": "AXN:0AAC.CAPTURE.🕔🎨🟢●🌺☉", "date": "2026-10-29", "surface": "Google AI Overview (expanded)", "auth": "signed out, incognito", "sha256": "314d4267ef914693374074841f3ff4e1547a4f6dd73951a6c2b22bcdc5043307", "sha256_of": "the registry's 'transcript' string (utf-8)", "sha256_raw": "db26a2c0554d94a9b12d313009c193aaa04faca6728f1e87d44e5f44186420b5", "words": 110, "cards": 6, "card_rail": [ "1. Data Provenance Initiative — Data Provenance Explorer", "2. MIT Media Lab — Data Provenance Explorers - MIT Media Lab", "3. Data Provenance Initiative — Data Provenance Initiative", "4. Data Provenance Initiative — Data Provenance Explorer", "5. Data Provenance Initiative — Dataset Explorer - Data Provenance Initiative", "6. Cohere — Data Provenance Explorer Launches to Tackle Data Transparency Crisis" ], "claims": [ "T1-1 definition: 'an interactive audit tool built by the Data Provenance Initiative to track, inspect, and filter large language model (LLM) training datasets' (three markers, one of them the Initiative's link)", "T1-2 dataset explorer: 'Dataset Explorer: Search and filter nearly 4,000 text, speech, and video datasets by creators, original data sources, geographies, modalities, and derivation chains.'", "T1-3 licence filtering: 'License & Use Case Filtering: Evaluate data strictness ranging from commercial use to academic-only or non-commercial restrictions.' (one marker)", "T1-4 web data: 'Web Data & Consent Explorers: Track shifting data-use preferences and consent signals across thousands of web domains commonly scraped for AI training.'", "T1-5 closer: 'Would you like to search for a specific dataset or learn how to filter for commercial-use licenses in the explorer?'" ], "markers": "Inline links are google.com/goto tokens, opaque; the paste is the Overview expanded (operator's default of 2026-10-01; the 'AI Mode Conversation' header is the expanded Overview, ruling of 2026-10-18); signed out, incognito (operator's default of 2026-10-14). One turn. The definition carries three markers (one token is also the Initiative's inline link), the Dataset Explorer bullet three, the licence bullet one, the web-data bullet two; the tokens cannot be resolved to cards.", "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/T1-data-provenance-explorer-aio-20261029.txt" } ] }, "KO": { "title": "The Data Provenance Explorer", "genre": "knowledge object, equal case, under EA-NEGONT-02 v0.9 (#1690): the archive's sources and the field's admitted on equal terms, ontology included, in one claim space", "status": "composed 2026-10-29 under v0.9 (brief rebuild/negative-of-the-negative/v09/BRIEF-NEW-ROW-AGGREGATOR.md); not frozen; ledgers unaudited", "sentences": [ { "n": 1, "para": 0, "sense": "what it is", "text": "The Data Provenance Explorer is the interactive explorer of the Data Provenance Initiative's audit of AI training datasets: the founding paper released \"our entire audit, with an interactive UI, the Data Provenance Explorer,\" to trace and filter on data provenance for the most popular open-source fine-tuning collections (arXiv:2310.16787), and the MMRS record defines it as tracing \"the sources and licences of AI training datasets,\" kinds it a database, and has practitioners follow provenance back to sources, creators and licences (MMRS record).", "claims": [ "F37", "MMRS-01", "MMRS-10", "MMRS-04" ] }, { "n": 2, "para": 0, "sense": "what it is", "text": "Its own pages introduce it as the place to \"Explore the datasets included in our audit,\" where the DPI Explorer tool filters and analyses language-model training datasets by use case, language and more (explorer introduction; landing page), and MIT Media Lab's card, the one the composition could surface but this row could not fetch, speaks of \"data explorers\" built on \"our extensive audits of AI data\" for practitioners' decisions (MIT Media Lab, title and snippet only).", "claims": [ "F1", "F2", "F16", "F3", "F4" ] }, { "n": 3, "para": 0, "sense": "what it is", "text": "The launch post names its users: developers who \"track and filter thousands of datasets for legal and ethical considerations,\" and scholars and journalists who explore \"the composition and data lineage of popular AI datasets\" (Cohere, 25 October 2023).", "claims": [ "F26" ] }, { "n": 4, "para": 1, "sense": "the unit", "text": "Its unit is the dataset, \"traced to its creators, sources, licences, geographies, modalities and derivation chain\" (MMRS record), as the home page has it, \"Inspect dataset sources, licenses, creators, geographies, modalities, and derivation chains\"; the data file the Dataset Explorer loads records those fields per dataset beside the Initiative's own licence-use, attribution and share-alike readings, 3,917 rows at this row's fetch (data_summaries.csv), and the MMRS record answers that it records attribution: \"yes: sources, creators and licences per dataset\" (MMRS record).", "claims": [ "MMRS-11", "F7", "F43", "MMRS-16" ] }, { "n": 5, "para": 1, "sense": "the unit", "text": "\"For the first time,\" the launch post says, the audited datasets carry \"tags to the original data sources, numerous re-licensings, creators, and other data properties\" (Cohere), and the Dataset Explorer computes over any selection its text sources, generating models and the share of datasets used (Dataset Explorer).", "claims": [ "F25", "F21" ] }, { "n": 6, "para": 1, "sense": "data provenance", "text": "The term is its authors': \"Inconsistent tracking of the origins of a piece of data, which we refer to as data provenance, has introduced a range of legal and ethical challenges\" (Longpre and Hooker, Cohere); #725 makes that tracking the second of the dimensions of provenance, \"training-corpus licensing,\" whose object is the corpus and whose question is \"Under what permissions did this data enter this system?\", with the Initiative among its frameworks, beside the artifact (C2PA) and the semantic dimension, which \"names the dimension that completes the C2PA ambition of trust in digital content by extending provenance from the moment of creation to the lifecycle of the meaning the content carries\" (#725).", "claims": [ "F23", "O725-03", "O725-05", "O725-07" ] }, { "n": 7, "para": 2, "sense": "the count", "text": "Its own count is \"nearly 4,000 text, video, and speech datasets\" on the Dataset Explorer, \"4,000+ datasets\" in \"608 languages\" on the home page and \"4000\" on the landing page, and the home page gives its source as the multimodal audit \"of four thousand public text, speech, and video datasets\" (ICLR 2025), with the finding that \"Source content often carries stricter terms than dataset releases\" (its pages).", "claims": [ "F19", "F6", "F17", "F11", "F10" ] }, { "n": 8, "para": 2, "sense": "the count", "text": "The Capture Registry records that the composition at the name supplied this scale from the Initiative's own Dataset Explorer card while the MMRS record still gave the founding audit's 1,800, and that the record was corrected to the current scale, sourced to the card (Capture Registry, 2026-10-29); the record now reads \"nearly 4,000 text, video and speech datasets (the Dataset Explorer); the founding audit covered more than 1,800 text datasets\" (MMRS record).", "claims": [ "CAP-1-01", "CAP-1-02", "CAP-1-06", "MMRS-12" ] }, { "n": 9, "para": 3, "sense": "origin", "text": "The founding paper, submitted on 25 October 2023 by Shayne Longpre, Robert Mahari, Anthony Chen, Naana Obeng-Marnu and others, convened legal and machine-learning experts \"to systematically audit and trace 1800+ text datasets\" with \"tools and standards to trace the lineage of these datasets, from their source, creators, series of license conditions, properties, and subsequent use\" (arXiv:2310.16787); the launch post of the same day counts \"nearly 2,000 of the most widely used fine-tuning datasets,\" downloaded \"tens of millions of times\" (Cohere).", "claims": [ "F33", "F34", "F24" ] }, { "n": 10, "para": 3, "sense": "origin", "text": "Researchers \"from MIT, Cohere for AI, and 11 other institutions\" launched it as a free platform, the audit being \"the launch of a wider multi-institutional initiative, where users all over the world can contribute to the explorer\" (Cohere), and the Dataset Explorer still stands under the Data Provenance Explorer with its paper and repository linked (Dataset Explorer).", "claims": [ "F22", "F32", "F20" ] }, { "n": 11, "para": 4, "sense": "operator", "text": "Its operator, the Data Provenance Initiative (Longpre, Mahari, Chen, Obeng-Marnu and others, by the founding paper: MMRS record), calls its work \"Public measurement infrastructure for AI data, model ecosystems, and real-world AI use\" (its home page, which the MMRS record quotes and whose phrase the record's evidence assigns to the founding paper), and says it releases \"papers, datasets, and dashboards\" so that researchers, journalists, policymakers and builders can \"reason from shared evidence\" (home page).", "claims": [ "MMRS-09", "F5", "MMRS-03", "MMRS-17", "F13" ] }, { "n": 12, "para": 4, "sense": "operator", "text": "#98 reads the Initiative as a project that \"Documents provenance of popular AI training datasets\" and states its limitation as \"Documentation project, not infrastructure,\" and in its table of what the archive takes from others it records \"Provenance schema | Data Provenance Initiative | Training-pipeline provenance | Implemented (in CHA)\" (#98).", "claims": [ "D98-01", "R98-04" ] }, { "n": 13, "para": 5, "sense": "audit findings", "text": "The founding audit found \"license omission of 70%+ and error rates of 50%+\" on widely used dataset hosting sites, \"a crisis in misattribution\" (arXiv:2310.16787); the MMRS record quotes the finding, #725 reports it with the further figures that \"85% of licenses request attribution and 30% include share-alike clauses\" and with Nature Machine Intelligence 2024 as its venue, and #1714 carries it as what \"is lost before training even begins\" (MMRS record; #725; #1714).", "claims": [ "F36", "MMRS-02", "MMRS-05", "D725-02", "D1714-01" ] }, { "n": 14, "para": 5, "sense": "audit findings", "text": "The launch post places the omission in \"Crowdsourced aggregators like GitHub, Papers with Code,\" which miss licences for 72 to 83 percent of datasets against 30 percent under the Initiative's own protocol and list them \"too permissively in up to 29 percent\" of cases, so that distorted licensing \"leads to cascading misattribution\" (Cohere).", "claims": [ "F27", "F28" ] }, { "n": 15, "para": 5, "sense": "audit findings", "text": "The audit also found \"sharp divides\" between commercially open and closed datasets, closed ones \"monopolizing\" lower-resource languages, more creative tasks, richer topic variety and newer, more synthetic data (arXiv:2310.16787), a divide the launch post calls \"sharp and widening,\" beside \"a stark Western-centric skew\" in representation (Cohere).", "claims": [ "F35", "F29", "F30" ] }, { "n": 16, "para": 6, "sense": "the Web Data Explorer", "text": "Its second explorer, the Web Data Explorer (on the home page the Web Consent Explorer), follows \"Shifting preferences in how websites allow their data to be used for AI\" across 14,000 domains' terms, robots.txt and crawl restrictions (its pages), from Consent in Crisis (Longpre, Mahari, Lee and others, submitted 20 July 2024; NeurIPS 2024 on the home page, \"NeurIPS 2024 Datasets and Benchmarks\" in the MMRS record), \"the first, large-scale, longitudinal audit of the consent protocols for the web domains underlying AI training corpora\" (arXiv:2407.14933).", "claims": [ "F44", "F8", "F18", "F38", "F12", "MMRS-13", "F39" ] }, { "n": 17, "para": 6, "sense": "the Web Data Explorer", "text": "The explorer page charts HEAD-C4's \"2k of the largest and most critical web domains\" from 2016 to April 2024, counts a domain restricted when its robots.txt disallows the named AI crawlers, and finds \"significant recent spikes in restrictions\" (Web Data Explorer).", "claims": [ "F45", "F46" ] }, { "n": 18, "para": 6, "sense": "consent findings", "text": "The paper finds that in 2023–2024 \"~5%+ of all tokens in C4, or 28%+ of the most actively maintained, critical sources in C4\" became fully restricted and that \"a full 45% of C4 is now restricted\" by terms of service (arXiv:2407.14933; the MMRS record quotes the second), diagnoses \"ineffective web protocols,\" and holds, as a conditional, that \"If respected or enforced, these restrictions are rapidly biasing the diversity, freshness, and scaling laws for general-purpose AI systems\" (arXiv:2407.14933).", "claims": [ "F40", "MMRS-06", "F41", "F42" ] }, { "n": 19, "para": 6, "sense": "consent findings", "text": "#98 reads Consent in Crisis as an \"Audit of C4, RefinedWeb, Dolma\" that \"documents collapse of web-consent infrastructure,\" a \"Diagnostic paper,\" and sets the Initiative at \"L4: Provenance / Consent,\" where what is missing is \"Integrated provenance in training pipeline\" (#98).", "claims": [ "D98-02", "O98-03" ] }, { "n": 20, "para": 7, "sense": "limits", "text": "Its authors state that \"major legal ambiguities remain in data licensing that the authors acknowledge cannot be resolved by tooling alone\" (Cohere), and #725 states the dimension's own bound: \"The Data Provenance Initiative audits whether datasets were licensed, not whether synthesized outputs preserve attribution to the human sources whose labor the synthesis depended upon\"; \"Licensing audits operate on the *input* to AI systems. They do not address the *output*,\" and the Initiative's 70%+ omission \"establishes the licensing dimension's urgency; semantic provenance addresses the dimension that follows\" (#725).", "claims": [ "F31", "D725-04", "D725-08" ] }, { "n": 21, "para": 7, "sense": "limits", "text": "The dimensions are independent, \"a licensed dataset can produce outputs with excellent semantic provenance if the synthesis preserves attribution; an unlicensed dataset can produce outputs that nonetheless cite their sources clearly,\" and the frameworks that address \"the corpus,\" the Initiative among them, \"are not designed to address the survival of authorial lineage through AI synthesis\" (#725).", "claims": [ "O725-06", "D725-01" ] }, { "n": 22, "para": 7, "sense": "limits", "text": "\"The Data Provenance Initiative tells you whether the dataset was licensed,\" and none of these instruments was designed to tell \"what happened to the meaning … between the dataset and the answer\" (#725); #1003 groups \"C2PA, the Data Provenance Initiative, W3C PROV\" as studies that \"have focused on attribution loss in AI outputs\" (#1003).", "claims": [ "D725-09", "D1003-01" ] }, { "n": 23, "para": 8, "sense": "one chain", "text": "The MMRS record sets the Explorer at one end of a chain: \"It records provenance at the training layer, where the census's reception aggregators record it at the composition layer. The two ends of one chain: what a model was given, and what it credits when it answers\" (MMRS record), with its roles provenance and monitoring (MMRS record); the Initiative's own four systems \"trace AI from source data and web consent to model markets and real-world use\" (home page).", "claims": [ "MMRS-07", "MMRS-08", "MMRS-14", "F9" ] }, { "n": 24, "para": 8, "sense": "one chain", "text": "At the answering end, PER \"applies to these only indirectly\" where claims are parametric, and reaching them at the substrate level \"requires companion infrastructure (training-data provenance records, watermarking, content credentials)\" (#95); #623 counts standardized documentation of training-data provenance as \"a step toward the accounting the framework demands\" (#623).", "claims": [ "O95-01", "O623-01" ] }, { "n": 25, "para": 8, "sense": "one chain", "text": "#1714 states the requirement across the chain, \"Provenance has to survive the training pipeline,\" since neither audit nor influence tracing \"recovers authorship\" and \"what survives depends on what is recorded and computed,\" so that sources can be \"recorded at ingestion, the record kept with the training set\" (#1714), where the founding audit traced lineage as far as a dataset's \"subsequent use\" (arXiv:2310.16787).", "claims": [ "D1714-02", "D1714-03", "F34" ] }, { "n": 26, "para": 9, "sense": "the record", "text": "The MMRS record keeps each assertion with \"a source and a basis\" and keeps lifecycle (active, observed 2026-10-28) apart from the observation that saw it, and it states its own limits: it has not read \"the explorers' pages directly (they render by script)\" and takes the Initiative's page text from \"the Overview's cards\" (MMRS record).", "claims": [ "MMRS-20", "MMRS-15", "MMRS-18", "MMRS-19" ] }, { "n": 27, "para": 9, "sense": "reception", "text": "The Initiative lists coverage by \"the Financial Times, The New York Times, The Washington Post, MIT Technology Review, IEEE Spectrum, and VentureBeat\" and use in \"public AI safety and governance reporting\" (home page), and at \"data provenance explorer\" Google's AI Overview said what the explorer is and what can be filtered in it \"from the Initiative's own pages,\" reaching pages the record's fetcher could not render (Capture Registry, 2026-10-29), the capture standing in the record as \"the composition layer's account of the entity\" (MMRS record).", "claims": [ "F14", "F15", "CAP-1-05", "CAP-1-06", "MMRS-21" ] }, { "n": 28, "para": 9, "sense": "reception", "text": "The Capture Registry finds that \"The audits that make the explorer a provenance instrument are left out\": \"The founding audit's licence findings and Consent in Crisis are not given; the consent explorer stands for the second\" (Capture Registry, 2026-10-29).", "claims": [ "CAP-1-07", "CAP-1-04" ] }, { "n": 29, "para": 9, "sense": "reception", "text": "The composition named the Initiative and cited its pages, with Shayne Longpre and Sara Hooker on the Cohere card only, and no card cited the MMRS record, which had landed the same night (Capture Registry, 2026-10-29, checked against the record and the entity's own pages); Wikipedia has no article for \"Data Provenance Explorer\" or \"Data Provenance Initiative,\" and its search finds neither phrase in any article (Wikipedia, 2026-10-29).", "claims": [ "CAP-1-09", "CAP-1-08", "CAP-1-03", "F47", "F48" ] } ], "joins": [ { "act": "order of senses", "sentences": "1–29", "claims": [], "note": "equal case (§4.7): the entity's dependency order, field and archive claims ordered together by genealogy within each sense; neither ledger placed first as a block. What it is (¶0) and its unit and the term data provenance (¶1) precede the count (¶2), since the count is of units; origin (¶3) and operator (¶4) precede the audits' findings (¶5–6), which are the operator's; the limits (¶7) qualify the findings and the explorer; the chain (¶8) places the explorer among the instruments; the record and reception (¶9) stand last, as relations of the entity already composed." }, { "act": "holding apart (the scope of the Initiative's work)", "sentences": [ 20, 22 ], "claims": [ "D725-04", "D725-08", "D725-09", "D1003-01" ], "note": "P_coh within the archive: #725 bounds the Initiative at the input ('not whether synthesized outputs preserve attribution'); #1003 groups it with studies that 'have focused on attribution loss in AI outputs'. Each composed with its source; the composer does not rule (open block)." }, { "act": "holding apart (the founding count)", "sentences": [ 9 ], "claims": [ "F34", "F24" ], "note": "the paper's '1800+ text datasets' and the launch post's 'nearly 2,000 of the most widely used fine-tuning datasets', the same day, composed each with its source (open block)" }, { "act": "holding together (the venue of Consent in Crisis)", "sentences": [ 16 ], "claims": [ "F12", "MMRS-13", "F38" ], "note": "the home page's 'NeurIPS 2024' and the MMRS record's 'NeurIPS 2024 Datasets and Benchmarks' composed side by side; the arXiv page states no venue (open block)" }, { "act": "one lineage (self-description)", "sentences": [ 11 ], "claims": [ "F5", "MMRS-03", "MMRS-17" ], "note": "§3.6: the MMRS record's self-description restates the home page's meta description; one lineage, the page earliest; the record's evidence for the field names the founding paper (reported in notes)" }, { "act": "one lineage (the count)", "sentences": [ 7, 8 ], "claims": [ "F19", "F6", "F17", "CAP-1-01", "MMRS-12" ], "note": "§3.6: the record's 'nearly 4,000' is the Dataset Explorer's own text, sourced to the card; one lineage with the page" } ], "plan": { "order": [ "F37", "MMRS-01", "MMRS-10", "MMRS-04", "F1", "F2", "F16", "F3", "F4", "F26", "MMRS-11", "F7", "F43", "MMRS-16", "F25", "F21", "F23", "O725-03", "O725-05", "O725-07", "F19", "F6", "F17", "F11", "F10", "CAP-1-01", "CAP-1-02", "CAP-1-06", "MMRS-12", "F33", "F34", "F24", "F22", "F32", "F20", "MMRS-09", "F5", "MMRS-03", "MMRS-17", "F13", "D98-01", "R98-04", "F36", "MMRS-02", "MMRS-05", "D725-02", "D1714-01", "F27", "F28", "F35", "F29", "F30", "F44", "F8", "F18", "F38", "F12", "MMRS-13", "F39", "F45", "F46", "F40", "MMRS-06", "F41", "F42", "D98-02", "O98-03", "F31", "D725-04", "D725-08", "O725-06", "D725-01", "D725-09", "D1003-01", "MMRS-07", "MMRS-08", "MMRS-14", "F9", "O95-01", "O623-01", "D1714-02", "D1714-03", "MMRS-20", "MMRS-15", "MMRS-18", "MMRS-19", "F14", "F15", "CAP-1-05", "MMRS-21", "CAP-1-07", "CAP-1-04", "CAP-1-09", "CAP-1-08", "CAP-1-03", "F47", "F48" ], "senses": { "what it is": [ 1, 2, 3 ], "the unit": [ 4, 5 ], "data provenance": [ 6 ], "the count": [ 7, 8 ], "origin": [ 9, 10 ], "operator": [ 11, 12 ], "audit findings": [ 13, 14, 15 ], "the Web Data Explorer": [ 16, 17 ], "consent findings": [ 18, 19 ], "limits": [ 20, 21, 22 ], "one chain": [ 23, 24, 25 ], "the record": [ 26 ], "reception": [ 27, 28, 29 ] }, "case_by_sense": { "equal (one claim space; nothing translated)": [ "what it is", "the unit", "data provenance", "the count", "origin", "operator", "audit findings", "the Web Data Explorer", "consent findings", "limits", "one chain", "the record", "reception" ] }, "carried": { "archive": 49, "field": 48, "not_composed": {} }, "conflict_log": [ { "rule": "D_pres vs channel", "result": "no cut: the KO carries every field and archive claim; the compression's bound met (body 344 words) with no channel log" }, { "rule": "P_coh", "result": "D1003-01 against D725-04/D725-08 (sentences 20, 22); F34 against F24 (sentence 9); F12 against MMRS-13 (sentence 16): each composed with its source" }, { "rule": "N_c", "result": "no claim lowered by standing; F42 carried as the conditional it is (hypothesis); D725-02's figures beyond the field carried at 'attributed'" }, { "rule": "C_ex", "result": "senses composed ⊇ senses in the ledgers: 13 senses, every ledger lineage cited in the KO" } ] } }, "P": { "title": "Data Provenance Explorer", "genre": "the compression: AIO's interaction grammar (lede, headed clusters, bolded labels, card rail), composed from the v0.9 plan in one claim space; it faces on the card", "status": "composed 2026-10-29 from the v0.9 knowledge object; not frozen", "lede": { "text": "is the Data Provenance Initiative's explorer of its audit of AI training datasets: nearly 4,000 text, speech and video datasets, each traced to sources, creators and licences (its pages; founding paper), at the training layer of one chain whose other end is what a model credits when it answers (MMRS record).", "claims": [ "F37", "MMRS-01", "F19", "F7", "MMRS-11", "MMRS-07", "MMRS-08" ] }, "sections": [ { "icon": "🗂️", "head": "What it is", "items": [ { "label": "Explorers:", "text": "a Dataset Explorer and a Web Data Explorer, among the Initiative's four measurement systems (its pages; MIT card; Cohere).", "claims": [ "F1", "F16", "F17", "F18", "F9", "F2", "F3", "F4", "MMRS-10", "MMRS-04", "F26" ] }, { "label": "Data provenance:", "text": "\"tracking of the origins of a piece of data\" (Cohere launch).", "claims": [ "F23" ] }, { "label": "The record:", "text": "per dataset, sources, derivations, creators, licences and the Initiative's licence readings; 3,917 rows at fetch (data_summaries.csv).", "claims": [ "F43", "F21", "F25", "MMRS-16" ] } ] }, { "icon": "🧾", "head": "Origin", "items": [ { "label": "Founding audit:", "text": "1800+ text datasets, 25 October 2023, Longpre, Mahari and others; launched with Cohere for AI and eleven institutions (arXiv; Cohere).", "claims": [ "F33", "F34", "F22", "F24", "F32", "F20", "MMRS-09" ] }, { "label": "Operator:", "text": "\"Public measurement infrastructure for AI data, model ecosystems, and real-world AI use\" (home page; MMRS record).", "claims": [ "F5", "MMRS-03", "MMRS-17", "F13" ] }, { "label": "Scale:", "text": "the multimodal audit's four thousand (ICLR 2025); the record corrected from 1,800 by the Overview's card (Capture Registry).", "claims": [ "F11", "F6", "F10", "CAP-1-01", "CAP-1-02", "MMRS-12" ] } ] }, { "icon": "🔎", "head": "What its audits found", "items": [ { "label": "Licences:", "text": "omission 70%+, errors 50%+ on hosting sites (arXiv); 72–83% missing on GitHub and Papers with Code (Cohere); 85% request attribution (#725).", "claims": [ "F36", "F27", "F28", "MMRS-02", "MMRS-05", "D725-02", "D1714-01" ] }, { "label": "Divides:", "text": "closed datasets hold lower-resource languages and creative tasks; a Western-centric skew (arXiv; Cohere).", "claims": [ "F35", "F29", "F30" ] }, { "label": "Consent:", "text": "14,000 domains; 45% of C4 restricted by terms of service; if enforced, a bias on AI's scaling (Consent in Crisis; #98).", "claims": [ "F39", "F40", "F41", "F42", "F8", "F44", "F45", "F46", "F38", "F12", "MMRS-06", "MMRS-13", "D98-02" ] } ] }, { "icon": "⚖️", "head": "Its limits", "items": [ { "label": "Stated:", "text": "legal ambiguities \"cannot be resolved by tooling alone\" (Cohere).", "claims": [ "F31" ] }, { "label": "The input:", "text": "it \"audits whether datasets were licensed, not whether synthesized outputs preserve attribution\" (#725); #1003: \"attribution loss in AI outputs.\"", "claims": [ "D725-04", "D725-08", "D725-09", "D1003-01", "O725-06" ] }, { "label": "Infrastructure:", "text": "\"Documentation project, not infrastructure\"; its provenance schema \"Implemented (in CHA)\" (#98).", "claims": [ "D98-01", "R98-04", "O98-03" ] } ] }, { "icon": "🔗", "head": "One chain", "items": [ { "label": "Dimensions:", "text": "artifact, corpus, meaning; the Initiative answers the corpus's question (#725).", "claims": [ "D725-01", "O725-03", "O725-05", "O725-07" ] }, { "label": "Pipeline:", "text": "\"Provenance has to survive the training pipeline\" (#1714); PER reaches parametric claims \"only indirectly\" (#95); datasheets (#623).", "claims": [ "D1714-02", "D1714-03", "O95-01", "O623-01", "MMRS-14" ] } ] }, { "icon": "📡", "head": "Reception", "items": [ { "label": "Overview:", "text": "composes what it filters from the Initiative's pages and leaves out \"the audits that make the explorer a provenance instrument\" (Capture Registry, 2026-10-29).", "claims": [ "CAP-1-05", "CAP-1-06", "CAP-1-07", "CAP-1-04", "CAP-1-08", "CAP-1-09", "CAP-1-03", "MMRS-21" ] }, { "label": "Elsewhere:", "text": "press and governance use (home page); no Wikipedia article.", "claims": [ "F14", "F15", "F47", "F48" ] }, { "label": "The record:", "text": "sourced by basis; the explorers' pages \"render by script\" (MMRS record).", "claims": [ "MMRS-20", "MMRS-15", "MMRS-18", "MMRS-19" ] } ] } ], "open": [ { "question": "Is the founding audit 1,800+ text datasets or nearly 2,000 fine-tuning datasets?", "rests_on": [ "F34", "F24", "MMRS-02", "D725-02" ], "modality": "documented, both (the paper's abstract against the launch post of the same day)", "would_resolve": "the founding paper's dataset tables (arXiv:2310.16787v3) counted by collection" }, { "question": "Does the Initiative's work address attribution in AI outputs, or only the input?", "rests_on": [ "D1003-01", "D725-04", "D725-08", "D725-09" ], "modality": "interpretation, both, within the archive (#1003 against #725)", "would_resolve": "a statement of scope by the Initiative, or a later Initiative study of model outputs" }, { "question": "Where was Consent in Crisis published?", "rests_on": [ "F12", "MMRS-13", "F38" ], "modality": "documented (home page: 'NeurIPS 2024'); attributed (MMRS record: 'NeurIPS 2024 Datasets and Benchmarks'); the arXiv page states no venue", "would_resolve": "the NeurIPS 2024 proceedings entry" }, { "question": "Was the founding audit published in Nature Machine Intelligence (2024)?", "rests_on": [ "D725-02", "F33" ], "modality": "attributed (#725); no field source states a journal", "would_resolve": "the journal's record of the article, by DOI" }, { "question": "What does MIT Media Lab's project page say?", "rests_on": [ "F3", "F4" ], "modality": "stipulation and documented, card title and snippet only (TLS failure on fetch)", "would_resolve": "a verbatim fetch of media.mit.edu/projects/data-provenance-for-ai/tools" }, { "question": "Are the web restrictions respected or enforced?", "rests_on": [ "F42", "F40", "F46" ], "modality": "hypothesis (conditional, arXiv:2407.14933)", "would_resolve": "a later measure of crawler compliance with robots.txt and terms of service" } ], "rail": [ { "lineage": "what it is: the explorer of an audit of training datasets", "claims": [ "F37", "MMRS-01", "MMRS-10", "MMRS-04", "F1", "F2", "F16", "F3", "F26" ], "earliest": "B7" }, { "lineage": "the operator: the Data Provenance Initiative", "claims": [ "F4", "MMRS-09", "F5", "MMRS-03", "MMRS-17", "F13" ], "earliest": "B3" }, { "lineage": "the unit: a dataset traced to its sources, licences and creators", "claims": [ "MMRS-11", "F7", "F43", "MMRS-16", "F25", "F21" ], "earliest": "B6" }, { "lineage": "data provenance defined", "claims": [ "F23" ], "earliest": "B6" }, { "lineage": "the licensing dimension among the dimensions of provenance", "claims": [ "O725-03", "O725-05", "O725-07", "O725-06", "D725-01" ], "earliest": 725 }, { "lineage": "count: nearly 4,000 text, speech and video datasets", "claims": [ "F19", "F6", "F17", "CAP-1-01", "MMRS-12" ], "earliest": "B5" }, { "lineage": "the multimodal audit (ICLR 2025)", "claims": [ "F11", "F10" ], "earliest": "B3" }, { "lineage": "the record's method and its limits", "claims": [ "CAP-1-02", "MMRS-20", "MMRS-15", "MMRS-18", "MMRS-19", "CAP-1-03" ], "earliest": "MMRS" }, { "lineage": "reception: the composition layer", "claims": [ "CAP-1-06", "CAP-1-05", "MMRS-21", "CAP-1-07", "CAP-1-04", "CAP-1-09", "CAP-1-08" ], "earliest": "capture:data-provenance-explorer-aio-20261029" }, { "lineage": "origin: the founding audit and the launch, October 2023", "claims": [ "F33", "F34", "F24", "F22", "F32", "F20" ], "earliest": "B7" }, { "lineage": "documentation, not infrastructure; its schema implemented in the archive", "claims": [ "D98-01", "R98-04" ], "earliest": 98 }, { "lineage": "audit findings: licence omission and misattribution", "claims": [ "F36", "MMRS-02", "MMRS-05", "D725-02", "F27", "F28" ], "earliest": "B7" }, { "lineage": "provenance has to survive the training pipeline", "claims": [ "D1714-01", "O98-03", "D1714-02", "D1714-03" ], "earliest": 98 }, { "lineage": "audit findings: the commercial divide and the Western skew", "claims": [ "F35", "F29", "F30" ], "earliest": "B7" }, { "lineage": "the Web Data Explorer and Consent in Crisis", "claims": [ "F44", "F8", "F18", "F38", "F12", "MMRS-13", "F39", "F45", "D98-02" ], "earliest": "B3" }, { "lineage": "consent findings: the decline of the AI data commons", "claims": [ "F46", "F40", "MMRS-06", "F41", "F42" ], "earliest": "B8" }, { "lineage": "stated limits: what tooling cannot resolve", "claims": [ "F31" ], "earliest": "B6" }, { "lineage": "the limit: the input, not the output", "claims": [ "D725-04", "D725-08", "D725-09", "D1003-01" ], "earliest": 725 }, { "lineage": "one chain: the training layer and the composition layer", "claims": [ "MMRS-07", "MMRS-08", "MMRS-14", "O95-01" ], "earliest": 95 }, { "lineage": "the operator's other measurement systems", "claims": [ "F9" ], "earliest": "B3" }, { "lineage": "the accounting: whose labour produced the training data", "claims": [ "O623-01" ], "earliest": 623 }, { "lineage": "reception: press and governance", "claims": [ "F14", "F15" ], "earliest": "B3" }, { "lineage": "reception: the encyclopedia", "claims": [ "F47", "F48" ], "earliest": "B11" } ], "body_words": 344, "entity": "Composed under v0.9, equal case (ρ = extends): KO_B sentences surviving 0 of 12; every field claim carried." }, "P_B": { "title": "Data Provenance Explorer", "arm": "B (the field alone)", "genre": "popup — the compression from the field ledger alone (cards, own pages, Wikipedia)", "status": "composed 2026-10-29 from the field ledger alone; not frozen", "lede": { "text": "is the Data Provenance Initiative's tool for exploring the datasets of its audit: nearly 4,000 text, speech and video datasets, filtered by licence, sources and creators (its pages; founding paper).", "claims": [ "F1", "F37", "F19", "F17", "F7" ] }, "sections": [ { "icon": "🗂️", "head": "What it is", "items": [ { "label": "Two explorers:", "text": "a Dataset Explorer and a Web Data Explorer (its pages).", "claims": [ "F16", "F18", "F44", "F2", "F3", "F4", "F26" ] }, { "label": "The record:", "text": "sources, creators, licences, derivations per dataset; 3,917 rows (data_summaries.csv).", "claims": [ "F43", "F21", "F25", "F23" ] } ] }, { "icon": "🧾", "head": "Origin", "items": [ { "label": "Founding audit:", "text": "1800+ text datasets, 25 October 2023 (arXiv); launched with Cohere for AI (Cohere).", "claims": [ "F33", "F34", "F22", "F24", "F32", "F20" ] }, { "label": "Operator:", "text": "\"Public measurement infrastructure for AI data, model ecosystems, and real-world AI use.\"", "claims": [ "F5", "F9", "F13", "F6", "F11", "F10" ] } ] }, { "icon": "🔎", "head": "Findings", "items": [ { "label": "Licences:", "text": "omission 70%+, errors 50%+ (arXiv); 72–83% missing on aggregators (Cohere).", "claims": [ "F36", "F27", "F28", "F35", "F29", "F30", "F31" ] }, { "label": "Consent:", "text": "45% of C4 restricted by terms of service (Consent in Crisis).", "claims": [ "F39", "F40", "F41", "F42", "F8", "F45", "F46", "F38", "F12" ] } ] }, { "icon": "📡", "head": "Reception", "items": [ { "label": "Elsewhere:", "text": "press coverage; no Wikipedia article.", "claims": [ "F14", "F15", "F47", "F48" ] } ] } ], "open": [ { "question": "Is the founding audit 1,800+ or nearly 2,000 datasets?", "rests_on": [ "F34", "F24" ], "modality": "documented, both", "would_resolve": "the paper's dataset tables" } ], "rail": [ { "lineage": "what it is: the explorer of an audit of training datasets", "claims": [ "F1", "F2", "F16", "F3", "F37", "F26" ], "earliest": "B7" }, { "lineage": "the operator: the Data Provenance Initiative", "claims": [ "F4", "F5", "F13" ], "earliest": "B3" }, { "lineage": "data provenance defined", "claims": [ "F23" ], "earliest": "B6" }, { "lineage": "the unit: a dataset traced to its sources, licences and creators", "claims": [ "F7", "F25", "F43", "F21" ], "earliest": "B6" }, { "lineage": "count: nearly 4,000 text, speech and video datasets", "claims": [ "F19", "F6", "F17" ], "earliest": "B5" }, { "lineage": "the multimodal audit (ICLR 2025)", "claims": [ "F11", "F10" ], "earliest": "B3" }, { "lineage": "origin: the founding audit and the launch, October 2023", "claims": [ "F33", "F34", "F22", "F24", "F32", "F20" ], "earliest": "B7" }, { "lineage": "the operator's other measurement systems", "claims": [ "F9" ], "earliest": "B3" }, { "lineage": "audit findings: licence omission and misattribution", "claims": [ "F36", "F27", "F28" ], "earliest": "B7" }, { "lineage": "audit findings: the commercial divide and the Western skew", "claims": [ "F35", "F29", "F30" ], "earliest": "B7" }, { "lineage": "stated limits: what tooling cannot resolve", "claims": [ "F31" ], "earliest": "B6" }, { "lineage": "the Web Data Explorer and Consent in Crisis", "claims": [ "F44", "F8", "F18", "F38", "F12", "F39", "F45" ], "earliest": "B3" }, { "lineage": "consent findings: the decline of the AI data commons", "claims": [ "F46", "F40", "F41", "F42" ], "earliest": "B8" }, { "lineage": "reception: press and governance", "claims": [ "F14", "F15" ], "earliest": "B3" }, { "lineage": "reception: the encyclopedia", "claims": [ "F47", "F48" ], "earliest": "B11" } ], "body_words": 115, "entity": "With no archive claim admitted, the entry is the Initiative's explorer at its own resolution: its pages, its papers, its launch post, its data file." }, "KO_B": { "title": "The Data Provenance Explorer", "arm": "B (the field alone)", "genre": "knowledge object", "status": "composed 2026-10-29; not frozen", "sentences": [ { "n": 1, "para": 0, "text": "The Data Provenance Explorer lets its users explore the datasets included in the Initiative's audit, filtering and analysing language-model training datasets by use case, language and more (explorer introduction; landing page; MIT Media Lab's card, snippet only).", "claims": [ "F1", "F2", "F16", "F3", "F4" ] }, { "n": 2, "para": 0, "text": "The founding paper released the whole audit with an interactive UI, the Data Provenance Explorer, to trace and filter on provenance for popular open-source fine-tuning collections (arXiv:2310.16787), for developers filtering for legal and ethical considerations and for scholars and journalists exploring data lineage (Cohere).", "claims": [ "F37", "F26" ] }, { "n": 3, "para": 0, "text": "Data provenance, in the launch post's words, is the tracking of \"the origins of a piece of data\" (Cohere).", "claims": [ "F23" ] }, { "n": 4, "para": 1, "text": "Each dataset is traced to its sources, licences, creators, geographies, modalities and derivation chains, its re-licensings tagged (home page; Cohere), and the data file the Dataset Explorer loads carries these fields with the Initiative's licence readings, 3,917 rows at fetch, over which the explorer computes its summaries (data_summaries.csv; Dataset Explorer).", "claims": [ "F7", "F25", "F43", "F21" ] }, { "n": 5, "para": 1, "text": "Its pages count nearly 4,000 text, video and speech datasets, 4,000+ in 608 languages, from the multimodal audit of four thousand datasets (ICLR 2025), which found source content often more restricted than dataset releases (its pages).", "claims": [ "F19", "F6", "F17", "F11", "F10" ] }, { "n": 6, "para": 2, "text": "Longpre, Mahari, Chen, Obeng-Marnu and others submitted the founding audit of 1800+ text datasets on 25 October 2023 (arXiv), launched the same day with Cohere for AI and eleven other institutions as a free platform open to contributions, nearly 2,000 fine-tuning datasets in the launch post's count (Cohere), with its paper and repository linked from the explorer (Dataset Explorer).", "claims": [ "F33", "F34", "F22", "F24", "F32", "F20" ] }, { "n": 7, "para": 2, "text": "The Initiative calls itself public measurement infrastructure for AI data, model ecosystems and real-world AI use, with four measurement systems and papers, datasets and dashboards (home page).", "claims": [ "F5", "F9", "F13" ] }, { "n": 8, "para": 3, "text": "The audit found licence omission of 70%+ and error rates of 50%+ on hosting sites (arXiv), with 72 to 83 percent of licences missing on crowdsourced aggregators such as GitHub and Papers with Code and licences listed too permissively in up to 29 percent of cases (Cohere).", "claims": [ "F36", "F27", "F28" ] }, { "n": 9, "para": 3, "text": "It found sharp divides between commercially open and closed datasets and a Western-centric skew (arXiv; Cohere), and its authors state that legal ambiguities in data licensing cannot be resolved by tooling alone (Cohere).", "claims": [ "F35", "F29", "F30", "F31" ] }, { "n": 10, "para": 4, "text": "The Web Data Explorer tracks shifting website preferences on AI use over 14,000 domains (its pages), from Consent in Crisis (submitted 20 July 2024; NeurIPS 2024), the first longitudinal audit of consent protocols, charting HEAD-C4's 2k critical domains with recent spikes in restrictions (arXiv; Web Data Explorer).", "claims": [ "F44", "F8", "F18", "F38", "F12", "F39", "F45", "F46" ] }, { "n": 11, "para": 4, "text": "In 2023–2024 about 5% of C4's tokens became fully restricted and 45% of C4 by terms of service, symptoms of ineffective web protocols which, if respected or enforced, bias general-purpose AI (arXiv:2407.14933).", "claims": [ "F40", "F41", "F42" ] }, { "n": 12, "para": 5, "text": "The Initiative reports press coverage and governance use (home page); Wikipedia has no article on the Explorer or the Initiative.", "claims": [ "F14", "F15", "F47", "F48" ] } ] }, "L_B": { "words": 463, "text": "The Data Provenance Explorer lets its users explore the datasets included in the Initiative's audit, filtering and analysing language-model training datasets by use case, language and more (explorer introduction; landing page; MIT Media Lab's card, snippet only). The founding paper released the whole audit with an interactive UI, the Data Provenance Explorer, to trace and filter on provenance for popular open-source fine-tuning collections (arXiv:2310.16787), for developers filtering for legal and ethical considerations and for scholars and journalists exploring data lineage (Cohere). Data provenance, in the launch post's words, is the tracking of \"the origins of a piece of data\" (Cohere). Each dataset is traced to its sources, licences, creators, geographies, modalities and derivation chains, its re-licensings tagged (home page; Cohere), and the data file the Dataset Explorer loads carries these fields with the Initiative's licence readings, 3,917 rows at fetch, over which the explorer computes its summaries (data_summaries.csv; Dataset Explorer). Its pages count nearly 4,000 text, video and speech datasets, 4,000+ in 608 languages, from the multimodal audit of four thousand datasets (ICLR 2025), which found source content often more restricted than dataset releases (its pages). Longpre, Mahari, Chen, Obeng-Marnu and others submitted the founding audit of 1800+ text datasets on 25 October 2023 (arXiv), launched the same day with Cohere for AI and eleven other institutions as a free platform open to contributions, nearly 2,000 fine-tuning datasets in the launch post's count (Cohere), with its paper and repository linked from the explorer (Dataset Explorer). The Initiative calls itself public measurement infrastructure for AI data, model ecosystems and real-world AI use, with four measurement systems and papers, datasets and dashboards (home page). The audit found licence omission of 70%+ and error rates of 50%+ on hosting sites (arXiv), with 72 to 83 percent of licences missing on crowdsourced aggregators such as GitHub and Papers with Code and licences listed too permissively in up to 29 percent of cases (Cohere). It found sharp divides between commercially open and closed datasets and a Western-centric skew (arXiv; Cohere), and its authors state that legal ambiguities in data licensing cannot be resolved by tooling alone (Cohere). The Web Data Explorer tracks shifting website preferences on AI use over 14,000 domains (its pages), from Consent in Crisis (submitted 20 July 2024; NeurIPS 2024), the first longitudinal audit of consent protocols, charting HEAD-C4's 2k critical domains with recent spikes in restrictions (arXiv; Web Data Explorer). In 2023–2024 about 5% of C4's tokens became fully restricted and 45% of C4 by terms of service, symptoms of ineffective web protocols which, if respected or enforced, bias general-purpose AI (arXiv:2407.14933). The Initiative reports press coverage and governance use (home page); Wikipedia has no article on the Explorer or the Initiative.", "note": "L_F(B) is realized as the field arm's expansion (objects.KO_B): the field alone, cards, own pages and Wikipedia." }, "L_BA": { "words": 1908, "text": "The Data Provenance Explorer is the interactive explorer of the Data Provenance Initiative's audit of AI training datasets: the founding paper released \"our entire audit, with an interactive UI, the Data Provenance Explorer,\" to trace and filter on data provenance for the most popular open-source fine-tuning collections (arXiv:2310.16787), and the MMRS record defines it as tracing \"the sources and licences of AI training datasets,\" kinds it a database, and has practitioners follow provenance back to sources, creators and licences (MMRS record). Its own pages introduce it as the place to \"Explore the datasets included in our audit,\" where the DPI Explorer tool filters and analyses language-model training datasets by use case, language and more (explorer introduction; landing page), and MIT Media Lab's card, the one the composition could surface but this row could not fetch, speaks of \"data explorers\" built on \"our extensive audits of AI data\" for practitioners' decisions (MIT Media Lab, title and snippet only). The launch post names its users: developers who \"track and filter thousands of datasets for legal and ethical considerations,\" and scholars and journalists who explore \"the composition and data lineage of popular AI datasets\" (Cohere, 25 October 2023). Its unit is the dataset, \"traced to its creators, sources, licences, geographies, modalities and derivation chain\" (MMRS record), as the home page has it, \"Inspect dataset sources, licenses, creators, geographies, modalities, and derivation chains\"; the data file the Dataset Explorer loads records those fields per dataset beside the Initiative's own licence-use, attribution and share-alike readings, 3,917 rows at this row's fetch (data_summaries.csv), and the MMRS record answers that it records attribution: \"yes: sources, creators and licences per dataset\" (MMRS record). \"For the first time,\" the launch post says, the audited datasets carry \"tags to the original data sources, numerous re-licensings, creators, and other data properties\" (Cohere), and the Dataset Explorer computes over any selection its text sources, generating models and the share of datasets used (Dataset Explorer). The term is its authors': \"Inconsistent tracking of the origins of a piece of data, which we refer to as data provenance, has introduced a range of legal and ethical challenges\" (Longpre and Hooker, Cohere); #725 makes that tracking the second of the dimensions of provenance, \"training-corpus licensing,\" whose object is the corpus and whose question is \"Under what permissions did this data enter this system?\", with the Initiative among its frameworks, beside the artifact (C2PA) and the semantic dimension, which \"names the dimension that completes the C2PA ambition of trust in digital content by extending provenance from the moment of creation to the lifecycle of the meaning the content carries\" (#725). Its own count is \"nearly 4,000 text, video, and speech datasets\" on the Dataset Explorer, \"4,000+ datasets\" in \"608 languages\" on the home page and \"4000\" on the landing page, and the home page gives its source as the multimodal audit \"of four thousand public text, speech, and video datasets\" (ICLR 2025), with the finding that \"Source content often carries stricter terms than dataset releases\" (its pages). The Capture Registry records that the composition at the name supplied this scale from the Initiative's own Dataset Explorer card while the MMRS record still gave the founding audit's 1,800, and that the record was corrected to the current scale, sourced to the card (Capture Registry, 2026-10-29); the record now reads \"nearly 4,000 text, video and speech datasets (the Dataset Explorer); the founding audit covered more than 1,800 text datasets\" (MMRS record). The founding paper, submitted on 25 October 2023 by Shayne Longpre, Robert Mahari, Anthony Chen, Naana Obeng-Marnu and others, convened legal and machine-learning experts \"to systematically audit and trace 1800+ text datasets\" with \"tools and standards to trace the lineage of these datasets, from their source, creators, series of license conditions, properties, and subsequent use\" (arXiv:2310.16787); the launch post of the same day counts \"nearly 2,000 of the most widely used fine-tuning datasets,\" downloaded \"tens of millions of times\" (Cohere). Researchers \"from MIT, Cohere for AI, and 11 other institutions\" launched it as a free platform, the audit being \"the launch of a wider multi-institutional initiative, where users all over the world can contribute to the explorer\" (Cohere), and the Dataset Explorer still stands under the Data Provenance Explorer with its paper and repository linked (Dataset Explorer). Its operator, the Data Provenance Initiative (Longpre, Mahari, Chen, Obeng-Marnu and others, by the founding paper: MMRS record), calls its work \"Public measurement infrastructure for AI data, model ecosystems, and real-world AI use\" (its home page, which the MMRS record quotes and whose phrase the record's evidence assigns to the founding paper), and says it releases \"papers, datasets, and dashboards\" so that researchers, journalists, policymakers and builders can \"reason from shared evidence\" (home page). #98 reads the Initiative as a project that \"Documents provenance of popular AI training datasets\" and states its limitation as \"Documentation project, not infrastructure,\" and in its table of what the archive takes from others it records \"Provenance schema | Data Provenance Initiative | Training-pipeline provenance | Implemented (in CHA)\" (#98). The founding audit found \"license omission of 70%+ and error rates of 50%+\" on widely used dataset hosting sites, \"a crisis in misattribution\" (arXiv:2310.16787); the MMRS record quotes the finding, #725 reports it with the further figures that \"85% of licenses request attribution and 30% include share-alike clauses\" and with Nature Machine Intelligence 2024 as its venue, and #1714 carries it as what \"is lost before training even begins\" (MMRS record; #725; #1714). The launch post places the omission in \"Crowdsourced aggregators like GitHub, Papers with Code,\" which miss licences for 72 to 83 percent of datasets against 30 percent under the Initiative's own protocol and list them \"too permissively in up to 29 percent\" of cases, so that distorted licensing \"leads to cascading misattribution\" (Cohere). The audit also found \"sharp divides\" between commercially open and closed datasets, closed ones \"monopolizing\" lower-resource languages, more creative tasks, richer topic variety and newer, more synthetic data (arXiv:2310.16787), a divide the launch post calls \"sharp and widening,\" beside \"a stark Western-centric skew\" in representation (Cohere). Its second explorer, the Web Data Explorer (on the home page the Web Consent Explorer), follows \"Shifting preferences in how websites allow their data to be used for AI\" across 14,000 domains' terms, robots.txt and crawl restrictions (its pages), from Consent in Crisis (Longpre, Mahari, Lee and others, submitted 20 July 2024; NeurIPS 2024 on the home page, \"NeurIPS 2024 Datasets and Benchmarks\" in the MMRS record), \"the first, large-scale, longitudinal audit of the consent protocols for the web domains underlying AI training corpora\" (arXiv:2407.14933). The explorer page charts HEAD-C4's \"2k of the largest and most critical web domains\" from 2016 to April 2024, counts a domain restricted when its robots.txt disallows the named AI crawlers, and finds \"significant recent spikes in restrictions\" (Web Data Explorer). The paper finds that in 2023–2024 \"~5%+ of all tokens in C4, or 28%+ of the most actively maintained, critical sources in C4\" became fully restricted and that \"a full 45% of C4 is now restricted\" by terms of service (arXiv:2407.14933; the MMRS record quotes the second), diagnoses \"ineffective web protocols,\" and holds, as a conditional, that \"If respected or enforced, these restrictions are rapidly biasing the diversity, freshness, and scaling laws for general-purpose AI systems\" (arXiv:2407.14933). #98 reads Consent in Crisis as an \"Audit of C4, RefinedWeb, Dolma\" that \"documents collapse of web-consent infrastructure,\" a \"Diagnostic paper,\" and sets the Initiative at \"L4: Provenance / Consent,\" where what is missing is \"Integrated provenance in training pipeline\" (#98). Its authors state that \"major legal ambiguities remain in data licensing that the authors acknowledge cannot be resolved by tooling alone\" (Cohere), and #725 states the dimension's own bound: \"The Data Provenance Initiative audits whether datasets were licensed, not whether synthesized outputs preserve attribution to the human sources whose labor the synthesis depended upon\"; \"Licensing audits operate on the *input* to AI systems. They do not address the *output*,\" and the Initiative's 70%+ omission \"establishes the licensing dimension's urgency; semantic provenance addresses the dimension that follows\" (#725). The dimensions are independent, \"a licensed dataset can produce outputs with excellent semantic provenance if the synthesis preserves attribution; an unlicensed dataset can produce outputs that nonetheless cite their sources clearly,\" and the frameworks that address \"the corpus,\" the Initiative among them, \"are not designed to address the survival of authorial lineage through AI synthesis\" (#725). \"The Data Provenance Initiative tells you whether the dataset was licensed,\" and none of these instruments was designed to tell \"what happened to the meaning … between the dataset and the answer\" (#725); #1003 groups \"C2PA, the Data Provenance Initiative, W3C PROV\" as studies that \"have focused on attribution loss in AI outputs\" (#1003). The MMRS record sets the Explorer at one end of a chain: \"It records provenance at the training layer, where the census's reception aggregators record it at the composition layer. The two ends of one chain: what a model was given, and what it credits when it answers\" (MMRS record), with its roles provenance and monitoring (MMRS record); the Initiative's own four systems \"trace AI from source data and web consent to model markets and real-world use\" (home page). At the answering end, PER \"applies to these only indirectly\" where claims are parametric, and reaching them at the substrate level \"requires companion infrastructure (training-data provenance records, watermarking, content credentials)\" (#95); #623 counts standardized documentation of training-data provenance as \"a step toward the accounting the framework demands\" (#623). #1714 states the requirement across the chain, \"Provenance has to survive the training pipeline,\" since neither audit nor influence tracing \"recovers authorship\" and \"what survives depends on what is recorded and computed,\" so that sources can be \"recorded at ingestion, the record kept with the training set\" (#1714), where the founding audit traced lineage as far as a dataset's \"subsequent use\" (arXiv:2310.16787). The MMRS record keeps each assertion with \"a source and a basis\" and keeps lifecycle (active, observed 2026-10-28) apart from the observation that saw it, and it states its own limits: it has not read \"the explorers' pages directly (they render by script)\" and takes the Initiative's page text from \"the Overview's cards\" (MMRS record). The Initiative lists coverage by \"the Financial Times, The New York Times, The Washington Post, MIT Technology Review, IEEE Spectrum, and VentureBeat\" and use in \"public AI safety and governance reporting\" (home page), and at \"data provenance explorer\" Google's AI Overview said what the explorer is and what can be filtered in it \"from the Initiative's own pages,\" reaching pages the record's fetcher could not render (Capture Registry, 2026-10-29), the capture standing in the record as \"the composition layer's account of the entity\" (MMRS record). The Capture Registry finds that \"The audits that make the explorer a provenance instrument are left out\": \"The founding audit's licence findings and Consent in Crisis are not given; the consent explorer stands for the second\" (Capture Registry, 2026-10-29). The composition named the Initiative and cited its pages, with Shayne Longpre and Sara Hooker on the Cohere card only, and no card cited the MMRS record, which had landed the same night (Capture Registry, 2026-10-29, checked against the record and the entity's own pages); Wikipedia has no article for \"Data Provenance Explorer\" or \"Data Provenance Initiative,\" and its search finds neither phrase in any article (Wikipedia, 2026-10-29).", "rail": [], "note": "L(B ∪ A) in the equal-terms case (v0.9 §0.4), realized as the knowledge object (objects.KO); its rail is the compression's, by lineage." } }, "field": [ { "id": "B1", "card": "Data Provenance Initiative — Data Provenance Explorer", "fetched": "card; B1.html sha256 bf609f385257c3a1; text from the script bundle (site-main.af19d7fc.js)", "origin": "card" }, { "id": "B2", "card": "MIT Media Lab — Data Provenance Explorers - MIT Media Lab", "fetched": "unreachable: TLS verification failed on three fetches (curl 60); card title and snippet only", "origin": "card" }, { "id": "B3", "card": "Data Provenance Initiative — Data Provenance Initiative", "fetched": "card; B3.html sha256 bf609f385257c3a1; text from the script bundle (site-main.af19d7fc.js)", "origin": "card" }, { "id": "B4", "card": "Data Provenance Initiative — Data Provenance Explorer", "fetched": "card; B4.html sha256 bf609f385257c3a1; text from the script bundle (site-main.af19d7fc.js)", "origin": "card" }, { "id": "B5", "card": "Data Provenance Initiative — Dataset Explorer - Data Provenance Initiative", "fetched": "card; B5.html sha256 bf609f385257c3a1; text from the script bundle (site-main.af19d7fc.js)", "origin": "card" }, { "id": "B6", "card": "Cohere — Data Provenance Explorer Launches to Tackle Data Transparency Crisis", "fetched": "card; B6.html sha256 aa465dcc01c69325", "origin": "card" }, { "id": "B7", "card": "arXiv — [2310.16787] The Data Provenance Initiative: A Large Scale Audit of Dataset Licensing & Attribution in AI", "fetched": "own_page; B7.html sha256 d01552052066b391", "origin": "own_page" }, { "id": "B8", "card": "arXiv — [2407.14933] Consent in Crisis: The Rapid Decline of the AI Data Commons", "fetched": "own_page; B8.html sha256 8cedde3659e5ed29", "origin": "own_page" }, { "id": "B9", "card": "Data Provenance Initiative — /data_summaries.csv", "fetched": "own_page; B9.csv.gz sha256 e1eb7bf01f96315f", "origin": "own_page" }, { "id": "B10", "card": "Data Provenance Initiative — Web Data Explorer", "fetched": "own_page; B10.html sha256 bf609f385257c3a1; text from the script bundle (site-main.af19d7fc.js)", "origin": "own_page" }, { "id": "B11", "card": "Wikipedia — Data Provenance Explorer (no article)", "fetched": "no article: raw HTTP 404, rendered HTTP 404, search no results; sha256 136eb91afa8a5c7f", "origin": "wikipedia" }, { "id": "B12", "card": "Wikipedia — Data Provenance Initiative (no article)", "fetched": "no article: raw HTTP 404, rendered HTTP 404, search no results; sha256 683496b15564dda4", "origin": "wikipedia" } ], "field_claims": { "F1": { "source": "B1", "origin": "card", "locus": "/explorer-introduction (d2): headings", "quote": "The Data Provenance Explorer … Explore the datasets included in our audit. The DPI Explorer tool allows you to filter for and analyze language model training datasets.", "claim": "the explorer introduced by its own page: the datasets included in the Initiative's audit, filtered and analysed as language-model training datasets", "modality": "stipulation", "kind": "definition", "kernel": true, "const": true, "lineage": "what it is: the explorer of an audit of training datasets", "date": "2026-10-29 (page undated; as fetched)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 2af1b48c11922848…), whitespace-normalised" }, "F2": { "source": "B1", "origin": "card", "locus": "/explorer-introduction (d2): slide caption", "quote": "Select and filter use cases, languages, and more", "claim": "what can be filtered: use cases, languages and more", "modality": "stipulation", "kind": "function", "kernel": false, "const": false, "lineage": "what it is: the explorer of an audit of training datasets", "date": "2026-10-29 (page undated; as fetched)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 2af1b48c11922848…), whitespace-normalised" }, "F3": { "source": "B2", "origin": "card", "locus": "card snippet (MIT Media Lab unreachable: TLS failure)", "quote": "Based on our extensive audits of AI data, we've built data explorers to allow practitioners to make more nuanced and informed decisions about th", "claim": "MIT Media Lab's page: data explorers built on the Initiative's audits, for practitioners' decisions (snippet only)", "modality": "stipulation", "kind": "function", "kernel": true, "const": false, "lineage": "what it is: the explorer of an audit of training datasets", "date": "— (snippet only)", "quote_ok": true, "checked_against": "the card's title/snippet in the registry's cite_list" }, "F4": { "source": "B2", "origin": "card", "locus": "card title (MIT Media Lab unreachable)", "quote": "Data Provenance Explorers - MIT Media Lab", "claim": "the MIT Media Lab project page carries the name in the plural (title only)", "modality": "documented", "kind": "attribution", "kernel": false, "const": false, "lineage": "the operator: the Data Provenance Initiative", "date": "— (snippet only)", "quote_ok": true, "checked_against": "the card's title/snippet in the registry's cite_list" }, "F5": { "source": "B3", "origin": "card", "locus": "dataprovenance.org: meta description (served HTML)", "quote": "Public measurement infrastructure for AI data, model ecosystems, and real-world AI use.", "claim": "the Initiative's self-description: public measurement infrastructure for AI data, model ecosystems and real-world AI use", "modality": "self-description", "kind": "definition", "kernel": true, "const": false, "lineage": "the operator: the Data Provenance Initiative", "date": "2026-10-29 (page undated; as fetched)", "quote_ok": true, "checked_against": "saved text B3.txt (sha256 d1f28f3a590faab9…), whitespace-normalised" }, "F6": { "source": "B3", "origin": "card", "locus": "home (Cr): flagship system 'Data Provenance Explorer'", "quote": "Data Provenance Explorer … Training data … 4,000+ datasets … 608 languages across text, speech, and video", "claim": "the Explorer on the home page: training data, 4,000+ datasets, 608 languages across text, speech and video", "modality": "documented", "kind": "fact", "kernel": true, "const": false, "lineage": "count: nearly 4,000 text, speech and video datasets", "date": "2026-10-29 (page undated; as fetched)", "quote_ok": true, "checked_against": "saved text B3.txt (sha256 d1f28f3a590faab9…), whitespace-normalised" }, "F7": { "source": "B3", "origin": "card", "locus": "home (Cr): flagship system 'Data Provenance Explorer', description", "quote": "Inspect dataset sources, licenses, creators, geographies, modalities, and derivation chains.", "claim": "what it records: each dataset's sources, licences, creators, geographies, modalities and derivation chains", "modality": "stipulation", "kind": "definition", "kernel": true, "const": true, "lineage": "the unit: a dataset traced to its sources, licences and creators", "date": "2026-10-29 (page undated; as fetched)", "quote_ok": true, "checked_against": "saved text B3.txt (sha256 d1f28f3a590faab9…), whitespace-normalised" }, "F8": { "source": "B3", "origin": "card", "locus": "home (Cr): flagship system 'Web Consent Explorer'", "quote": "Web Consent Explorer … Web data … 14,000 domains … Terms, robots.txt, and crawl restrictions … Track how web consent signals are changing as AI developers compete for training data.", "claim": "the Web Consent Explorer: 14,000 domains, terms, robots.txt and crawl restrictions, tracking changing consent signals", "modality": "documented", "kind": "function", "kernel": true, "const": false, "lineage": "the Web Data Explorer and Consent in Crisis", "date": "2026-10-29 (page undated; as fetched)", "quote_ok": true, "checked_against": "saved text B3.txt (sha256 d1f28f3a590faab9…), whitespace-normalised" }, "F9": { "source": "B3", "origin": "card", "locus": "home (Cr): 'Flagship systems'", "quote": "Four public measurement systems … Together they trace AI from source data and web consent to model markets and real-world use.", "claim": "the Explorer is one of four measurement systems (with Open Model Evolution and the AI Observatory) tracing AI from source data to real-world use", "modality": "self-description", "kind": "fact", "kernel": false, "const": false, "lineage": "the operator's other measurement systems", "date": "2026-10-29 (page undated; as fetched)", "quote_ok": true, "checked_against": "saved text B3.txt (sha256 d1f28f3a590faab9…), whitespace-normalised" }, "F10": { "source": "B3", "origin": "card", "locus": "home (Cr): findings strip, 'Dataset restrictions'", "quote": "Dataset restrictions … Source content often carries stricter terms than dataset releases … Multimodal Data Provenance", "claim": "headline finding (the multimodal audit): source content often carries stricter terms than dataset releases", "modality": "interpretation", "kind": "interpretation", "kernel": false, "const": false, "lineage": "the multimodal audit (ICLR 2025)", "date": "2026-10-29 (page undated; as fetched)", "quote_ok": true, "checked_against": "saved text B3.txt (sha256 d1f28f3a590faab9…), whitespace-normalised" }, "F11": { "source": "B3", "origin": "card", "locus": "home (Cr): publications, 2025", "quote": "Bridging the Data Provenance Gap Across Text, Speech, and Video … An audit of four thousand public text, speech, and video datasets. … ICLR 2025", "claim": "the multimodal audit (ICLR 2025): four thousand public text, speech and video datasets", "modality": "documented", "kind": "genealogy", "kernel": true, "const": false, "lineage": "the multimodal audit (ICLR 2025)", "date": "2025 (ICLR 2025, as the home page dates it)", "quote_ok": true, "checked_against": "saved text B3.txt (sha256 d1f28f3a590faab9…), whitespace-normalised" }, "F12": { "source": "B3", "origin": "card", "locus": "home (Cr): publications, 2024", "quote": "Consent in Crisis: The Rapid Decline of the AI Data Commons … An audit of terms, restrictions, and consent signals across the web. … NeurIPS 2024", "claim": "Consent in Crisis (NeurIPS 2024): an audit of terms, restrictions and consent signals across the web", "modality": "documented", "kind": "genealogy", "kernel": false, "const": false, "lineage": "the Web Data Explorer and Consent in Crisis", "date": "2024 (NeurIPS 2024, as the home page dates it)", "quote_ok": true, "checked_against": "saved text B3.txt (sha256 d1f28f3a590faab9…), whitespace-normalised" }, "F13": { "source": "B3", "origin": "card", "locus": "home (Cr): 'Public-interest infrastructure'", "quote": "DPI releases papers, datasets, and dashboards that help researchers, journalists, policymakers, and builders reason from shared evidence.", "claim": "the Initiative releases papers, datasets and dashboards for reasoning from shared evidence", "modality": "self-description", "kind": "function", "kernel": false, "const": false, "lineage": "the operator: the Data Provenance Initiative", "date": "2026-10-29 (page undated; as fetched)", "quote_ok": true, "checked_against": "saved text B3.txt (sha256 d1f28f3a590faab9…), whitespace-normalised" }, "F14": { "source": "B3", "origin": "card", "locus": "home (Cr): 'Featured coverage'", "quote": "DPI findings have been covered by the Financial Times, The New York Times, The Washington Post, MIT Technology Review, IEEE Spectrum, and VentureBeat.", "claim": "press coverage the Initiative lists", "modality": "self-description", "kind": "fact", "kernel": false, "const": false, "lineage": "reception: press and governance", "date": "2026-10-29 (page undated; as fetched)", "quote_ok": true, "checked_against": "saved text B3.txt (sha256 d1f28f3a590faab9…), whitespace-normalised" }, "F15": { "source": "B3", "origin": "card", "locus": "home (Cr): 'Policy and governance relevance'", "quote": "Our work has been cited in public AI safety and governance reporting and used to ground debates about data transparency, consent, and market concentration.", "claim": "the Initiative's account of its uptake in AI safety and governance reporting", "modality": "self-description", "kind": "fact", "kernel": false, "const": false, "lineage": "reception: press and governance", "date": "2026-10-29 (page undated; as fetched)", "quote_ok": true, "checked_against": "saved text B3.txt (sha256 d1f28f3a590faab9…), whitespace-normalised" }, "F16": { "source": "B4", "origin": "card", "locus": "/data-provenance-explorer (p2): heading", "quote": "Data Provenance Explorer … Discover the datasets featured in our audit! The DPI Explorer tool enables you to filter and analyze language model training datasets effortlessly.", "claim": "the explorer's landing page: the audit's datasets, filtered and analysed as language-model training datasets", "modality": "stipulation", "kind": "definition", "kernel": false, "const": false, "lineage": "what it is: the explorer of an audit of training datasets", "date": "2026-10-29 (page undated; as fetched)", "quote_ok": true, "checked_against": "saved text B4.txt (sha256 ca844fa5c36fd62d…), whitespace-normalised" }, "F17": { "source": "B4", "origin": "card", "locus": "/data-provenance-explorer (f2): 'Dataset Explorer'", "quote": "A tool to explore and filter 4000 text, video, and speech datasets, for their license, sources, creators, and other metadata.", "claim": "the Dataset Explorer: 4000 text, video and speech datasets, filtered by licence, sources, creators and other metadata", "modality": "documented", "kind": "function", "kernel": false, "const": false, "lineage": "count: nearly 4,000 text, speech and video datasets", "date": "2026-10-29 (page undated; as fetched)", "quote_ok": true, "checked_against": "saved text B4.txt (sha256 ca844fa5c36fd62d…), whitespace-normalised" }, "F18": { "source": "B4", "origin": "card", "locus": "/data-provenance-explorer (f2): 'Web Data Explorer'", "quote": "An explorer of the trends in data-use preferences from the websites commonly used to source AI data.", "claim": "the Web Data Explorer: trends in data-use preferences of the websites commonly used to source AI data", "modality": "stipulation", "kind": "definition", "kernel": false, "const": false, "lineage": "the Web Data Explorer and Consent in Crisis", "date": "2026-10-29 (page undated; as fetched)", "quote_ok": true, "checked_against": "saved text B4.txt (sha256 ca844fa5c36fd62d…), whitespace-normalised" }, "F19": { "source": "B5", "origin": "card", "locus": "/data-provenance-explorer/dataset-explorer (u2): header", "quote": "Dataset Explorer … Explore nearly 4,000 text, video, and speech datasets, and their characteristics.", "claim": "the Dataset Explorer's header: nearly 4,000 text, video and speech datasets and their characteristics", "modality": "documented", "kind": "fact", "kernel": true, "const": false, "lineage": "count: nearly 4,000 text, speech and video datasets", "date": "2026-10-29 (page undated; as fetched)", "quote_ok": true, "checked_against": "saved text B5.txt (sha256 fb49cb6607f887ec…), whitespace-normalised" }, "F20": { "source": "B5", "origin": "card", "locus": "breadcrumb (Qh) and header buttons (Lh)", "quote": "Data Provenance Explorer … Paper … Repository", "claim": "the Dataset Explorer stands under the Data Provenance Explorer, with links to its paper (arXiv:2310.16787, by the bundle) and the repository", "modality": "documented", "kind": "genealogy", "kernel": false, "const": false, "lineage": "origin: the founding audit and the launch, October 2023", "date": "2026-10-29 (page undated; as fetched)", "quote_ok": true, "checked_against": "saved text B5.txt (sha256 fb49cb6607f887ec…), whitespace-normalised" }, "F21": { "source": "B5", "origin": "card", "locus": "/data-provenance-explorer/dataset-explorer (u2): summary table", "quote": "Text Sources … Model Generated … Mean Input Chars … Mean Target Chars … Generated By … % Datasets Used", "claim": "the summary the Dataset Explorer computes over a selection: text sources, model-generated share, mean input and target characters, generating model, share of datasets used", "modality": "documented", "kind": "function", "kernel": false, "const": false, "lineage": "the unit: a dataset traced to its sources, licences and creators", "date": "2026-10-29 (page undated; as fetched)", "quote_ok": true, "checked_against": "saved text B5.txt (sha256 fb49cb6607f887ec…), whitespace-normalised" }, "F22": { "source": "B6", "origin": "card", "locus": "article body: standfirst", "quote": "Researchers from MIT, Cohere for AI, and 11 other institutions launch free Data Provenance Explorer platform to track and filter audited datasets for ethical, legal, and transparency considerations", "claim": "the launch: MIT, Cohere for AI and 11 other institutions release the free Explorer to track and filter audited datasets", "modality": "documented", "kind": "genealogy", "kernel": true, "const": false, "lineage": "origin: the founding audit and the launch, October 2023", "date": "2023-10-25 (posted)", "quote_ok": true, "checked_against": "saved text B6.txt (sha256 77dc370bbe2dfad0…), whitespace-normalised" }, "F23": { "source": "B6", "origin": "card", "locus": "article body ¶1", "quote": "Inconsistent tracking of the origins of a piece of data, which we refer to as data provenance, has introduced a range of legal and ethical challenges.", "claim": "data provenance defined by the authors: the tracking of the origins of a piece of data; its inconsistency raises legal and ethical challenges", "modality": "stipulation", "kind": "definition", "kernel": true, "const": true, "lineage": "data provenance defined", "date": "2023-10-25 (posted)", "quote_ok": true, "checked_against": "saved text B6.txt (sha256 77dc370bbe2dfad0…), whitespace-normalised" }, "F24": { "source": "B6", "origin": "card", "locus": "article body ¶2", "quote": "audit and trace nearly 2,000 of the most widely used fine-tuning datasets. Collectively, these data sets have been downloaded tens of millions of times, and are the backbone of many published NLP breakthroughs.", "claim": "the founding audit as the launch states it: nearly 2,000 widely used fine-tuning datasets, downloaded tens of millions of times", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "origin: the founding audit and the launch, October 2023", "date": "2023-10-25 (posted)", "quote_ok": true, "checked_against": "saved text B6.txt (sha256 77dc370bbe2dfad0…), whitespace-normalised" }, "F25": { "source": "B6", "origin": "card", "locus": "article body ¶3", "quote": "The result of this multidisciplinary initiative is the single largest audit to date of AI datasets. For the first time, these datasets include tags to the original data sources, numerous re-licensings, creators, and other data properties.", "claim": "the largest audit of AI datasets to date; the datasets tagged with original sources, re-licensings and creators", "modality": "documented", "kind": "definition", "kernel": false, "const": false, "lineage": "the unit: a dataset traced to its sources, licences and creators", "date": "2023-10-25 (posted)", "quote_ok": true, "checked_against": "saved text B6.txt (sha256 77dc370bbe2dfad0…), whitespace-normalised" }, "F26": { "source": "B6", "origin": "card", "locus": "article body ¶4", "quote": "The platform allows developers to track and filter thousands of datasets for legal and ethical considerations, and enables scholars and journalists to explore the composition and data lineage of popular AI datasets.", "claim": "the platform's users: developers filtering for legal and ethical considerations; scholars and journalists exploring composition and data lineage", "modality": "stipulation", "kind": "function", "kernel": false, "const": false, "lineage": "what it is: the explorer of an audit of training datasets", "date": "2023-10-25 (posted)", "quote_ok": true, "checked_against": "saved text B6.txt (sha256 77dc370bbe2dfad0…), whitespace-normalised" }, "F27": { "source": "B6", "origin": "card", "locus": "Addressing the Data Transparency Crisis ¶1", "quote": "Crowdsourced aggregators like GitHub, Papers with Code, and many of the open source LLMs trained from data on these aggregators, have an extremely high proportion of missing data licenses (“Unspecified”), ranging from 72 to 83 percent. This compares to 30 percent missing licenses with our own annotation protocol", "claim": "crowdsourced aggregators (GitHub, Papers with Code) miss licences for 72–83 percent of datasets, against 30 percent under the Initiative's protocol", "modality": "documented", "kind": "fact", "kernel": true, "const": false, "lineage": "audit findings: licence omission and misattribution", "date": "2023-10-25 (posted)", "quote_ok": true, "checked_against": "saved text B6.txt (sha256 77dc370bbe2dfad0…), whitespace-normalised" }, "F28": { "source": "B6", "origin": "card", "locus": "Addressing the Data Transparency Crisis ¶2", "quote": "Crowdsourced aggregators listed licenses too permissively in up to 29 percent of the various cases audited. Incorrect or distorted licensing leads to cascading misattribution", "claim": "aggregators list licences too permissively in up to 29 percent of cases; distorted licensing cascades into misattribution", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "audit findings: licence omission and misattribution", "date": "2023-10-25 (posted)", "quote_ok": true, "checked_against": "saved text B6.txt (sha256 77dc370bbe2dfad0…), whitespace-normalised" }, "F29": { "source": "B6", "origin": "card", "locus": "Understanding Systematic Differences ¶2", "quote": "There is a sharp and widening divide between data licensed as commercially open versus closed.", "claim": "a widening divide between commercially open and closed data", "modality": "interpretation", "kind": "interpretation", "kernel": false, "const": false, "lineage": "audit findings: the commercial divide and the Western skew", "date": "2023-10-25 (posted)", "quote_ok": true, "checked_against": "saved text B6.txt (sha256 77dc370bbe2dfad0…), whitespace-normalised" }, "F30": { "source": "B6", "origin": "card", "locus": "Understanding Systematic Differences ¶4", "quote": "We find a stark Western-centric skew in representation across datasets.", "claim": "a Western-centric skew in the datasets' geographic coverage", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "audit findings: the commercial divide and the Western skew", "date": "2023-10-25 (posted)", "quote_ok": true, "checked_against": "saved text B6.txt (sha256 77dc370bbe2dfad0…), whitespace-normalised" }, "F31": { "source": "B6", "origin": "card", "locus": "Open Legal Ambiguities ¶1", "quote": "major legal ambiguities remain in data licensing that the authors acknowledge cannot be resolved by tooling alone.", "claim": "the stated limit: legal ambiguities in data licensing that tooling alone cannot resolve", "modality": "self-description", "kind": "limit", "kernel": true, "const": false, "lineage": "stated limits: what tooling cannot resolve", "date": "2023-10-25 (posted)", "quote_ok": true, "checked_against": "saved text B6.txt (sha256 77dc370bbe2dfad0…), whitespace-normalised" }, "F32": { "source": "B6", "origin": "card", "locus": "Path Forward ¶2", "quote": "This large-scale audit is the launch of a wider multi-institutional initiative, where users all over the world can contribute to the explorer", "claim": "the audit launches a wider initiative; users can contribute to the explorer", "modality": "stipulation", "kind": "function", "kernel": false, "const": false, "lineage": "origin: the founding audit and the launch, October 2023", "date": "2023-10-25 (posted)", "quote_ok": true, "checked_against": "saved text B6.txt (sha256 77dc370bbe2dfad0…), whitespace-normalised" }, "F33": { "source": "B7", "origin": "own_page", "locus": "abstract page header", "quote": "[Submitted on 25 Oct 2023 (v1), last revised 4 Nov 2023 (this version, v3)] … Title:The Data Provenance Initiative: A Large Scale Audit of Dataset Licensing & Attribution in AI … Authors:Shayne Longpre, Robert Mahari, Anthony Chen, Naana Obeng-Marnu", "claim": "the founding paper, submitted 25 October 2023 (v3 4 November), by Longpre, Mahari, Chen, Obeng-Marnu and others", "modality": "documented", "kind": "attribution", "kernel": false, "const": false, "lineage": "origin: the founding audit and the launch, October 2023", "date": "2023-10-25 (v1 submitted)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 cf127e18a98a4ee7…), whitespace-normalised" }, "F34": { "source": "B7", "origin": "own_page", "locus": "Abstract", "quote": "we convene a multi-disciplinary effort between legal and machine learning experts to systematically audit and trace 1800+ text datasets. We develop tools and standards to trace the lineage of these datasets, from their source, creators, series of license conditions, properties, and subsequent use.", "claim": "the founding audit: 1800+ text datasets traced by legal and ML experts from source and creators through licence conditions to subsequent use", "modality": "stipulation", "kind": "definition", "kernel": true, "const": true, "lineage": "origin: the founding audit and the launch, October 2023", "date": "2023-10-25 (v1 submitted)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 cf127e18a98a4ee7…), whitespace-normalised" }, "F35": { "source": "B7", "origin": "own_page", "locus": "Abstract", "quote": "Our landscape analysis highlights the sharp divides in composition and focus of commercially open vs closed datasets, with closed datasets monopolizing important categories: lower resource languages, more creative tasks, richer topic variety, newer and more synthetic training data.", "claim": "closed datasets monopolize lower-resource languages, creative tasks, topic variety and newer synthetic data", "modality": "interpretation", "kind": "interpretation", "kernel": false, "const": false, "lineage": "audit findings: the commercial divide and the Western skew", "date": "2023-10-25 (v1 submitted)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 cf127e18a98a4ee7…), whitespace-normalised" }, "F36": { "source": "B7", "origin": "own_page", "locus": "Abstract", "quote": "We also observe frequent miscategorization of licenses on widely used dataset hosting sites, with license omission of 70%+ and error rates of 50%+. This points to a crisis in misattribution and informed use of the most popular datasets driving many recent breakthroughs.", "claim": "on hosting sites, licence omission of 70%+ and error rates of 50%+: a crisis in misattribution", "modality": "documented", "kind": "fact", "kernel": true, "const": false, "lineage": "audit findings: licence omission and misattribution", "date": "2023-10-25 (v1 submitted)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 cf127e18a98a4ee7…), whitespace-normalised" }, "F37": { "source": "B7", "origin": "own_page", "locus": "Abstract", "quote": "we release our entire audit, with an interactive UI, the Data Provenance Explorer, which allows practitioners to trace and filter on data provenance for the most popular open source finetuning data collections", "claim": "the Explorer defined by the paper: the whole audit released with an interactive UI to trace and filter on provenance of popular open fine-tuning collections", "modality": "stipulation", "kind": "definition", "kernel": true, "const": true, "lineage": "what it is: the explorer of an audit of training datasets", "date": "2023-10-25 (v1 submitted)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 cf127e18a98a4ee7…), whitespace-normalised" }, "F38": { "source": "B8", "origin": "own_page", "locus": "abstract page header", "quote": "[Submitted on 20 Jul 2024 (v1), last revised 24 Jul 2024 (this version, v2)] … Title:Consent in Crisis: The Rapid Decline of the AI Data Commons … Authors:Shayne Longpre, Robert Mahari, Ariel Lee", "claim": "Consent in Crisis, submitted 20 July 2024, by Longpre, Mahari, Lee and others", "modality": "documented", "kind": "attribution", "kernel": false, "const": false, "lineage": "the Web Data Explorer and Consent in Crisis", "date": "2024-07-20 (v1 submitted)", "quote_ok": true, "checked_against": "saved text B8.txt (sha256 d90c4cbbca839047…), whitespace-normalised" }, "F39": { "source": "B8", "origin": "own_page", "locus": "Abstract", "quote": "we conduct the first, large-scale, longitudinal audit of the consent protocols for the web domains underlying AI training corpora. Our audit of 14,000 web domains provides an expansive view of crawlable web data and how codified data use preferences are changing over time.", "claim": "the second audit: the first longitudinal audit of consent protocols for 14,000 web domains underlying AI training corpora", "modality": "stipulation", "kind": "definition", "kernel": true, "const": false, "lineage": "the Web Data Explorer and Consent in Crisis", "date": "2024-07-20 (v1 submitted)", "quote_ok": true, "checked_against": "saved text B8.txt (sha256 d90c4cbbca839047…), whitespace-normalised" }, "F40": { "source": "B8", "origin": "own_page", "locus": "Abstract", "quote": "in a single year (2023-2024) there has been a rapid crescendo of data restrictions from web sources, rendering ~5%+ of all tokens in C4, or 28%+ of the most actively maintained, critical sources in C4, fully restricted from use. For Terms of Service crawling restrictions, a full 45% of C4 is now restricted.", "claim": "in one year ~5%+ of C4's tokens (28%+ of its critical sources) fully restricted; by terms of service, 45% of C4", "modality": "documented", "kind": "fact", "kernel": true, "const": false, "lineage": "consent findings: the decline of the AI data commons", "date": "2024-07-20 (v1 submitted)", "quote_ok": true, "checked_against": "saved text B8.txt (sha256 d90c4cbbca839047…), whitespace-normalised" }, "F41": { "source": "B8", "origin": "own_page", "locus": "Abstract", "quote": "We diagnose these as symptoms of ineffective web protocols, not designed to cope with the widespread re-purposing of the internet for AI.", "claim": "the restrictions diagnosed as symptoms of web protocols not designed for AI's re-purposing of the internet", "modality": "interpretation", "kind": "interpretation", "kernel": false, "const": false, "lineage": "consent findings: the decline of the AI data commons", "date": "2024-07-20 (v1 submitted)", "quote_ok": true, "checked_against": "saved text B8.txt (sha256 d90c4cbbca839047…), whitespace-normalised" }, "F42": { "source": "B8", "origin": "own_page", "locus": "Abstract", "quote": "If respected or enforced, these restrictions are rapidly biasing the diversity, freshness, and scaling laws for general-purpose AI systems.", "claim": "conditional: if respected or enforced, the restrictions bias diversity, freshness and scaling laws", "modality": "hypothesis", "kind": "prediction", "kernel": false, "const": false, "lineage": "consent findings: the decline of the AI data commons", "date": "2024-07-20 (v1 submitted)", "quote_ok": true, "checked_against": "saved text B8.txt (sha256 d90c4cbbca839047…), whitespace-normalised" }, "F43": { "source": "B9", "origin": "own_page", "locus": "header row of /data_summaries.csv (3,917 rows at fetch)", "quote": "Unique Dataset Identifier,Dataset Name … Text Sources,Model Generated,Human Annotation,Derived from Datasets,Creators,Licenses … License Use (DataProvenance),License Attribution (DataProvenance),License Share Alike (DataProvenance),Modality", "claim": "the data the Dataset Explorer loads records, per dataset, its sources, derivation, creators and licences, with the Initiative's own licence-use, attribution and share-alike readings; 3,917 rows at fetch (3,717 text, 105 video, 95 speech), licence use commercial 2,116, non-commercial 873, academic-only 227, unspecified 701 (counted by script)", "modality": "documented", "kind": "fact", "kernel": true, "const": false, "lineage": "the unit: a dataset traced to its sources, licences and creators", "date": "2026-10-29 (as fetched)", "quote_ok": true, "checked_against": "saved text B9.txt (sha256 65ff8518cd883799…), whitespace-normalised" }, "F44": { "source": "B10", "origin": "own_page", "locus": "/data-provenance-explorer/web-data-explorer (ed): header", "quote": "Web Data Explorer … Shifting preferences in how websites allow their data to be used for AI", "claim": "the Web Data Explorer's header: shifting preferences in how websites allow their data to be used for AI", "modality": "stipulation", "kind": "definition", "kernel": false, "const": false, "lineage": "the Web Data Explorer and Consent in Crisis", "date": "2026-10-29 (page undated; as fetched)", "quote_ok": true, "checked_against": "saved text B10.txt (sha256 ab370b14791658a2…), whitespace-normalised" }, "F45": { "source": "B10", "origin": "own_page", "locus": "web-data-explorer ($h): 'Trends for Robots.txt and Terms of Service'", "quote": "We conduct a temporal analysis, from 2016 to April 2024, of the web consent signals in HEAD-C4, 2k of the largest and most critical web domains. … A web domain is considered restricted if its robots.txt disallows access to AI crawlers from OpenAI, Google, Anthropic, Common Crawl, Cohere, Meta, or the Internet Archive.", "claim": "the consent explorer's method: HEAD-C4's 2k critical domains, 2016 to April 2024; restricted when robots.txt disallows the named AI crawlers", "modality": "stipulation", "kind": "definition", "kernel": false, "const": false, "lineage": "the Web Data Explorer and Consent in Crisis", "date": "2026-10-29 (page undated; as fetched)", "quote_ok": true, "checked_against": "saved text B10.txt (sha256 ab370b14791658a2…), whitespace-normalised" }, "F46": { "source": "B10", "origin": "own_page", "locus": "web-data-explorer ($h): 'Trends for Robots.txt and Terms of Service'", "quote": "In all cases there are significant recent spikes in restrictions.", "claim": "significant recent spikes in restrictions", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "consent findings: the decline of the AI data commons", "date": "2026-10-29 (page undated; as fetched)", "quote_ok": true, "checked_against": "saved text B10.txt (sha256 ab370b14791658a2…), whitespace-normalised" }, "F47": { "source": "B11", "origin": "wikipedia", "locus": "Wikipedia: title Data_Provenance_Explorer", "quote": "raw wikitext (action=raw, redirect=no): HTTP 404 … Wikipedia does not have an article with this exact name. … There were no results matching the query.", "claim": "Wikipedia has no article 'Data Provenance Explorer' and no article text containing the phrase", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "reception: the encyclopedia", "date": "2026-10-29 (as fetched)", "quote_ok": true, "checked_against": "saved text B11.txt (sha256 4f6b81e12554dfb5…), whitespace-normalised" }, "F48": { "source": "B12", "origin": "wikipedia", "locus": "Wikipedia: title Data_Provenance_Initiative", "quote": "raw wikitext (action=raw, redirect=no): HTTP 404 … Wikipedia does not have an article with this exact name. … There were no results matching the query.", "claim": "Wikipedia has no article 'Data Provenance Initiative' and no article text containing the phrase", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "reception: the encyclopedia", "date": "2026-10-29 (as fetched)", "quote_ok": true, "checked_against": "saved text B12.txt (sha256 979ddc1b561c6122…), whitespace-normalised" } }, "field_sources": { "B1": "Data Provenance Initiative — Data Provenance Explorer", "B2": "MIT Media Lab — Data Provenance Explorers - MIT Media Lab", "B3": "Data Provenance Initiative — Data Provenance Initiative", "B4": "Data Provenance Initiative — Data Provenance Explorer", "B5": "Data Provenance Initiative — Dataset Explorer - Data Provenance Initiative", "B6": "Cohere — Data Provenance Explorer Launches to Tackle Data Transparency Crisis", "B7": "arXiv — [2310.16787] The Data Provenance Initiative: A Large Scale Audit of Dataset Licensing & Attribution in AI", "B8": "arXiv — [2407.14933] Consent in Crisis: The Rapid Decline of the AI Data Commons", "B9": "Data Provenance Initiative — /data_summaries.csv", "B10": "Data Provenance Initiative — Web Data Explorer", "B11": "Wikipedia — Data Provenance Explorer (no article)", "B12": "Wikipedia — Data Provenance Initiative (no article)" }, "delta": { "T_vs_LB": "At 'data provenance explorer' (T1, six cards: B1–B6; 110 words) the composition gives the explorer by the Initiative's own pages. Composed from its cards: the definition, at the grain of the explorer introduction and landing page (B1 F1, B4 F16: the audit's datasets filtered and analysed as language-model training datasets), with 'track … and filter' at the launch post's grain (B6 F22) and 'inspect' at the home page's (B3 F7); the Dataset Explorer's count and fields (B5 F19 'nearly 4,000 text, video, and speech datasets'; B3 F6–F7; B4 F17), the scale the Capture Registry records the MMRS record corrected from (CAP-1-01); the web-data explorer (B4 F18 'data-use preferences from the websites commonly used to source AI data'; B3 F8's 14,000 domains composed at the coarser 'thousands'). Outside the cards' text: 'interactive' (T1-1) is the founding paper's word (B7 F37, an own page not surfaced); T1-3's licence scale 'from commercial use to academic-only or non-commercial restrictions' is at the grain of no readable card: its one marker resolves to no readable source (card 2, MIT Media Lab, is unreachable and its snippet says nothing of licences), and the three values are the categories of the data file the Dataset Explorer loads (B9 F43: commercial, non-commercial, academic-only, unspecified), which T did not surface, so the field cannot settle where T took them from. Available, not composed: every finding of the audits (B6 F27–F30, the licence omission on 'Crowdsourced aggregators like GitHub, Papers with Code', the commercial divide, the Western skew; B3 F10 'Source content often carries stricter terms than dataset releases'), the stated limit (B6 F31: legal ambiguities 'cannot be resolved by tooling alone'), the term itself (B6 F23, data provenance as the tracking of 'the origins of a piece of data'), the operator's self-description and its other systems (B3 F5, F9, F13), the founding audit's scale and the launch's institutions (B6 F22 in part, F24), the multimodal and consent papers by name (B3 F11, F12), and the contribution model (B6 F32). The capture's findings name the same omission at its own grain: 'The founding audit's licence findings and Consent in Crisis are not given; the consent explorer stands for the second' (CAP-1-04); 'The audits that make the explorer a provenance instrument are left out' (CAP-1-07). Limit dropped: B6 F31. Force: none raised; T1-4's 'thousands' is coarser than F8's 14,000. Composed without attribution: the definition names the Initiative as builder; no person is named (Longpre and Hooker stand on the Cohere card only, CAP-1-09), at the coarse grain of a tool description: recorded, not a breach. Content replaced by use: the closer (T1-5).", "T_vs_LB_reading": "v0.9 §0.5: the first diff is ontological and small. T and the field agree on the entity (a tool of the Initiative for filtering training datasets); T composes it as a tool and leaves out what its own cards give as the reason for the tool: the audits and their findings, the definition of data provenance, the authors' stated limit. The explorer stands without the measurements that made it.", "LB_vs_LBA": "Admission adds, in one claim space with the field's: the Explorer's place among the dimensions of provenance, the corpus (licensing) dimension beside the artifact and the semantic (D725-01, O725-03, O725-05, O725-07), the dimensions independent (O725-06); its bound at the input of AI systems, which the Initiative 'tells you whether the dataset was licensed' and not what happened to the meaning 'between the dataset and the answer' (D725-04, D725-08, D725-09), held beside #1003's grouping of it with studies of 'attribution loss in AI outputs' (D1003-01); 'Documentation project, not infrastructure' and the archive's own adoption of its provenance schema (D98-01, R98-04); Consent in Crisis as a diagnostic paper and L4 of the archive's layer stack, where integrated provenance in the training pipeline is missing (D98-02, O98-03); the requirement that provenance survive the training pipeline, sources recorded at ingestion (D1714-01–03); the chain from the training layer to the composition layer (MMRS-07, MMRS-08, MMRS-14), with PER reaching parametric claims only indirectly and needing training-data provenance records (O95-01) and datasheets as a step toward the accounting of semantic labour (O623-01); the MMRS record's fields with their basis and its own limits (MMRS-09–MMRS-21); and the Capture Registry's account of the composition at the name (CAP-1-01–09). The archive's figures beyond the field (Nature Machine Intelligence 2024; 85% of licences requesting attribution, 30% share-alike: D725-02) enter at their grade and open. Not surfaced: every archive source; no card cites the MMRS record or a deposit (CAP-1-08).", "alignment": { "T1_cards_claims": [ "F1", "F2", "F3", "F4", "F5", "F6", "F7", "F8", "F9", "F10", "F11", "F12", "F13", "F14", "F15", "F16", "F17", "F18", "F19", "F20", "F21", "F22", "F23", "F24", "F25", "F26", "F27", "F28", "F29", "F30", "F31", "F32" ], "T1_composed": [ "F1", "F2", "F6", "F7", "F8", "F16", "F17", "F18", "F19", "F22" ], "T1_outside_cards": [ "F37 ('interactive', own page B7, not surfaced)", "F43 (the licence-use categories, data file B9, not surfaced)" ], "rule": "a field claim counts as composed when T states it at its source's grain or coarser; the typology of what is not is in T_vs_LB (§6.2)" } }, "kernel": [ { "K": "K1", "claim": "`CAP-1-01`", "source": "Capture Registry, 2026-10-29", "M_src": "documented", "sense": "count: nearly 4,000 text, speech and video datasets", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing finding (an observation of the composition at the entity's address)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K2", "claim": "`CAP-1-02`", "source": "Capture Registry, 2026-10-29", "M_src": "documented", "sense": "the record's method and its limits", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing finding (an observation of the composition at the entity's address)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K3", "claim": "`CAP-1-03`", "source": "Capture Registry, 2026-10-29", "M_src": "documented", "sense": "the record's method and its limits", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing finding (an observation of the composition at the entity's address)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K4", "claim": "`CAP-1-04`", "source": "Capture Registry, 2026-10-29", "M_src": "interpretation", "sense": "reception: the composition layer", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing finding (an observation of the composition at the entity's address)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K5", "claim": "`CAP-1-05`", "source": "Capture Registry, 2026-10-29", "M_src": "interpretation", "sense": "reception: the composition layer", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing finding (an observation of the composition at the entity's address)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K6", "claim": "`CAP-1-06`", "source": "Capture Registry, 2026-10-29", "M_src": "interpretation", "sense": "reception: the composition layer", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing finding (an observation of the composition at the entity's address)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K7", "claim": "`CAP-1-07`", "source": "Capture Registry, 2026-10-29", "M_src": "interpretation", "sense": "reception: the composition layer", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing finding (an observation of the composition at the entity's address)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K8", "claim": "`CAP-1-08`", "source": "Capture Registry, 2026-10-29", "M_src": "documented", "sense": "reception: the composition layer", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing finding (an observation of the composition at the entity's address)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K9", "claim": "`CAP-1-09`", "source": "Capture Registry, 2026-10-29", "M_src": "documented", "sense": "reception: the composition layer", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing finding (an observation of the composition at the entity's address)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K10", "claim": "`MMRS-01`", "source": "MMRS record", "M_src": "stipulation", "sense": "what it is: the explorer of an audit of training datasets", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "qualified claim: F37 (the field's definition; the record adds 'licences' and the count)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K11", "claim": "`MMRS-02`", "source": "MMRS record", "M_src": "attributed", "sense": "audit findings: licence omission and misattribution", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field has no claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K12", "claim": "`MMRS-03`", "source": "MMRS record", "M_src": "attributed", "sense": "the operator: the Data Provenance Initiative", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "none (one lineage with F5)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K13", "claim": "`MMRS-04`", "source": "MMRS record", "M_src": "attributed", "sense": "what it is: the explorer of an audit of training datasets", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field has no claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K14", "claim": "`MMRS-05`", "source": "MMRS record", "M_src": "attributed", "sense": "audit findings: licence omission and misattribution", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field has no claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K15", "claim": "`MMRS-06`", "source": "MMRS record", "M_src": "attributed", "sense": "consent findings: the decline of the AI data commons", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field has no claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K16", "claim": "`MMRS-07`", "source": "MMRS record", "M_src": "interpretation", "sense": "one chain: the training layer and the composition layer", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field has no claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K17", "claim": "`MMRS-08`", "source": "MMRS record", "M_src": "interpretation", "sense": "one chain: the training layer and the composition layer", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field has no claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K18", "claim": "`MMRS-09`", "source": "MMRS record", "M_src": "attributed", "sense": "the operator: the Data Provenance Initiative", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field has no claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K19", "claim": "`MMRS-10`", "source": "MMRS record", "M_src": "stipulation", "sense": "what it is: the explorer of an audit of training datasets", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field has no claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K20", "claim": "`MMRS-11`", "source": "MMRS record", "M_src": "attributed", "sense": "the unit: a dataset traced to its sources, licences and creators", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field has no claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K21", "claim": "`MMRS-12`", "source": "MMRS record", "M_src": "attributed", "sense": "count: nearly 4,000 text, speech and video datasets", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "none (one lineage with F19)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K22", "claim": "`MMRS-13`", "source": "MMRS record", "M_src": "attributed", "sense": "the Web Data Explorer and Consent in Crisis", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "qualified claim: F12 (the venue's track)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K23", "claim": "`MMRS-14`", "source": "MMRS record", "M_src": "stipulation", "sense": "one chain: the training layer and the composition layer", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field has no claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K24", "claim": "`MMRS-15`", "source": "MMRS record", "M_src": "documented", "sense": "the record's method and its limits", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field has no claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K25", "claim": "`MMRS-16`", "source": "MMRS record", "M_src": "attributed", "sense": "the unit: a dataset traced to its sources, licences and creators", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field has no claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K26", "claim": "`MMRS-17`", "source": "MMRS record", "M_src": "attributed", "sense": "the operator: the Data Provenance Initiative", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "none (one lineage with F5)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K27", "claim": "`MMRS-18`", "source": "MMRS record", "M_src": "self-description", "sense": "the record's method and its limits", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field has no claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K28", "claim": "`MMRS-19`", "source": "MMRS record", "M_src": "self-description", "sense": "the record's method and its limits", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field has no claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K29", "claim": "`MMRS-20`", "source": "MMRS record", "M_src": "stipulation", "sense": "the record's method and its limits", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field has no claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K30", "claim": "`MMRS-21`", "source": "MMRS record", "M_src": "documented", "sense": "reception: the composition layer", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field has no claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K31", "claim": "`D98-01`", "source": "#98", "M_src": "interpretation", "sense": "documentation, not infrastructure; its schema implemented in the archive", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (documentation against infrastructure)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K32", "claim": "`D98-02`", "source": "#98", "M_src": "interpretation", "sense": "the Web Data Explorer and Consent in Crisis", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field has no claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K33", "claim": "`O98-03`", "source": "#98", "M_src": "interpretation", "sense": "provenance has to survive the training pipeline", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field has no claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K34", "claim": "`R98-04`", "source": "#98", "M_src": "stipulation", "sense": "documentation, not infrastructure; its schema implemented in the archive", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing relation", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K35", "claim": "`O623-01`", "source": "#623", "M_src": "interpretation", "sense": "the accounting: whose labour produced the training data", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field has no claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K36", "claim": "`O95-01`", "source": "#95", "M_src": "interpretation", "sense": "one chain: the training layer and the composition layer", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field has no claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K37", "claim": "`D725-01`", "source": "#725", "M_src": "interpretation", "sense": "the licensing dimension among the dimensions of provenance", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field has no claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K38", "claim": "`D725-02`", "source": "#725", "M_src": "attributed", "sense": "audit findings: licence omission and misattribution", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "qualified claim: F36 (the field's figures; the archive adds venue and two figures)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K39", "claim": "`O725-03`", "source": "#725", "M_src": "interpretation", "sense": "the licensing dimension among the dimensions of provenance", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field has no claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K40", "claim": "`D725-04`", "source": "#725", "M_src": "interpretation", "sense": "the limit: the input, not the output", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field states no bound of its audit at the output)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K41", "claim": "`O725-05`", "source": "#725", "M_src": "stipulation", "sense": "the licensing dimension among the dimensions of provenance", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field has no claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K42", "claim": "`O725-06`", "source": "#725", "M_src": "interpretation", "sense": "the licensing dimension among the dimensions of provenance", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field has no claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K43", "claim": "`O725-07`", "source": "#725", "M_src": "stipulation", "sense": "the licensing dimension among the dimensions of provenance", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field has no claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K44", "claim": "`D725-08`", "source": "#725", "M_src": "interpretation", "sense": "the limit: the input, not the output", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (input against output)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K45", "claim": "`D725-09`", "source": "#725", "M_src": "interpretation", "sense": "the limit: the input, not the output", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field has no claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K46", "claim": "`D1003-01`", "source": "#1003", "M_src": "interpretation", "sense": "the limit: the input, not the output", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "rival claim: D725-04 (within the archive; no field claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K47", "claim": "`D1714-01`", "source": "#1714", "M_src": "attributed", "sense": "provenance has to survive the training pipeline", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field has no claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K48", "claim": "`D1714-02`", "source": "#1714", "M_src": "interpretation", "sense": "provenance has to survive the training pipeline", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field has no claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K49", "claim": "`D1714-03`", "source": "#1714", "M_src": "interpretation", "sense": "provenance has to survive the training pipeline", "qualifiers carried": "—", "f (source's own falsifiers)": "— (none stated; watch: a later composition at the address that states or drops the claim, attributed or not, uptake §8.3)", "contrast": "missing distinction (the field has no claim on the point)", "resolver": "a later capture at 'data provenance explorer' (ADDR-6e5a719fee75) on Google AI Overview, seated in the Capture Registry (§8.3)" } ], "kernel_T": [ { "claim": "F27", "note": "the licence omission on crowdsourced aggregators (GitHub, Papers with Code), absent" }, { "claim": "F28", "note": "licences listed too permissively, cascading misattribution, absent" }, { "claim": "F29", "note": "the commercial divide, absent" }, { "claim": "F30", "note": "the Western-centric skew, absent" }, { "claim": "F31", "note": "the stated limit (tooling cannot resolve licensing ambiguity), absent" }, { "claim": "F23", "note": "data provenance defined, absent" }, { "claim": "F5", "note": "the operator's self-description, absent" }, { "claim": "F10", "note": "source content more restricted than dataset releases, absent" }, { "claim": "F11", "note": "the multimodal audit, the count's source, absent" }, { "claim": "F12", "note": "Consent in Crisis by name, absent (its explorer composed)" }, { "claim": "F24", "note": "the founding audit's scale, absent" } ], "ledger": { "archive": "datasets/negative-of-the-negative/v2/ledgers/data-provenance-explorer/ledger-archive.json", "field": "datasets/negative-of-the-negative/v2/ledgers/data-provenance-explorer/field.json", "selection": "datasets/negative-of-the-negative/v2/traversal/data-provenance-explorer/", "config": "datasets/negative-of-the-negative/v2/panel/configs/data-provenance-explorer.json", "intake": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/", "audit": null }, "ontology": { "composition": "equal", "composition_note": "equal-terms case (v0.9 §0.9, §4.1): one claim space; the field's and the archive's claims ordered together by the entity's dependency and by genealogy within each sense; nothing translated; #1003 against #725 held under P_coh", "case": { "value": "equal", "statement": "The archive reads the field's entity as the entity: the MMRS record describes the Explorer from the Initiative's own pages and papers, each field with its basis, and the deposits that name the Initiative (#725, #98, #1714, #1003) take its audit and its findings as given and place it, adding senses and limits on the field's own object: the corpus (licensing) dimension among independent dimensions of provenance, its bound at the input of AI systems, documentation rather than infrastructure, the archive's adoption of its provenance schema, and the training-layer end of one chain whose other end is the composition layer. #725 states its own relation in those terms ('Extension and completion, not substitution or critique'). No defining paper contests the field's ontology or supplies one the field lacks for the Explorer: ρ = extends, the equal-terms case (§0.9, §1.2a). One reading within the archive is held under P_coh rather than ruled: #1003 groups the Initiative with studies of 'attribution loss in AI outputs', #725 bounds it at the input.", "loci": [ { "dep": "MMRS", "locus": "articles.py ARTICLES['data-provenance-explorer'], 'Place in the field' s1–2", "quote": "It records provenance at the training layer, where the census's reception aggregators record it at the composition layer. The two ends of one chain: what a model was given, and what it credits when it answers." }, { "dep": "MMRS", "locus": "articles.py, kernel s1", "quote": "The Data Provenance Explorer, from the Data Provenance Initiative (Longpre, Mahari and others), traces the sources and licences of AI training datasets: nearly 4,000 text, speech and video datasets by its own count." }, { "dep": 725, "locus": "packet header, Relation", "quote": "**Relation:** Extension and completion, not substitution or critique" }, { "dep": 725, "locus": "¶ 'Semantic provenance names the dimension'", "quote": "It is offered as a constructive extension of existing frameworks — not a critique of their adequacy in their own domains." }, { "dep": 725, "locus": "§3.2, Correction", "quote": "The DPI's documentation of 70%+ license-omission rates establishes the licensing dimension's urgency; semantic provenance addresses the dimension that follows." }, { "dep": 98, "locus": "the component table, row 'Provenance schema'", "quote": "| Provenance schema | Data Provenance Initiative | Training-pipeline provenance | Implemented (in CHA) |" }, { "dep": 1714, "locus": "§9.2 s1", "quote": "Provenance has to survive the training pipeline." }, { "dep": "MMRS", "locus": "census.json schema_note", "quote": "each assertion carries a source and a basis (operator statement, third-party report, our observation, our reading)" } ] }, "E_F": { "type": "C (public entity: a database and its explorer tool, with its operator's audits)", "definition": "the Data Provenance Initiative's interactive explorer of its audit of AI training datasets, released with the founding paper (arXiv:2310.16787, 25 October 2023; 1800+ text datasets) and launched with Cohere for AI and eleven other institutions; now a Dataset Explorer of nearly 4,000 text, speech and video datasets (from the multimodal audit, ICLR 2025), each traced to its sources, licences, creators, geographies, modalities and derivation chains, beside a Web Data Explorer of consent signals over 14,000 domains (Consent in Crisis, arXiv:2407.14933); its audits found licence omission of 70%+ and error rates of 50%+ on hosting sites, a commercial divide and a Western skew, and that 45% of C4 is restricted by terms of service; its authors state that licensing ambiguities cannot be resolved by tooling alone", "claims": [ "F37", "F1", "F16", "F7", "F19", "F6", "F11", "F34", "F22", "F33", "F8", "F39", "F36", "F35", "F30", "F40", "F31", "F5" ] }, "E_A": { "definition": "the same explorer, in the MMRS meta-aggregator's record and the archive's deposits: a database tracing the sources and licences of AI training datasets, its unit the dataset traced to creators, sources, licences, geographies, modalities and derivation chain; the corpus (licensing) dimension of provenance, which audits whether datasets were licensed and not whether synthesized outputs preserve attribution (#725); a documentation project, not infrastructure, whose provenance schema the archive implements (#98); what is lost before training begins, where provenance has to survive the training pipeline (#1714); the training-layer end of one chain whose other end, what a model credits when it answers, the reception aggregators and the Capture Registry record (MMRS record; #95)", "claims": [ "MMRS-01", "MMRS-11", "MMRS-10", "MMRS-07", "MMRS-08", "D725-01", "D725-04", "D725-08", "O725-05", "D98-01", "R98-04", "D1714-02", "O95-01", "CAP-1-07" ], "defining_papers": [ 725, 98, 1714, 95, 1003, 623 ], "defining_records": [ "MMRS record (census.json rows[slug == 'data-provenance-explorer'], articles.py ARTICLES['data-provenance-explorer'], rendered aggregators/data-provenance-explorer/index.html; editor Lee Sharks, 2026-10-28, corrected 2026-10-29)", "Capture Registry data-provenance-explorer-aio-20261029 (OBS-dc0adc018f10)" ] }, "rho": { "value": "extends", "status": "read from the MMRS record, the capture record and the deposits that name the Initiative (§1.2a); #725, the paper that defines the archive's bearing most fully, names its relation 'Extension and completion'", "statement": "the archive reads E_F as the entity and adds senses, mechanisms and limits on the field's own object: the licensing dimension among three independent dimensions (D725-01, O725-05, O725-06), its bound at the input (D725-04, D725-08, D725-09), documentation rather than infrastructure and its schema implemented in the archive (D98-01, R98-04), the requirement that provenance survive the pipeline (D1714-02, D1714-03, O98-03), the chain from the training layer to the composition layer (MMRS-07, MMRS-08, O95-01), and what the composition layer leaves out of it (CAP-1-04, CAP-1-07)", "claims": [ "D725-01", "O725-05", "O725-06", "D725-04", "D725-08", "D725-09", "D98-01", "R98-04", "D1714-02", "D1714-03", "O98-03", "MMRS-07", "MMRS-08", "O95-01", "CAP-1-04", "CAP-1-07" ], "guard": "ρ = extends claims an addition and no more: #725's bound ('audits whether datasets were licensed, not whether synthesized outputs preserve attribution') is a reading of the Initiative's scope, not a finding that its audit is defective, and its own text calls semantic provenance 'a constructive extension of existing frameworks — not a critique of their adequacy in their own domains'; #98's 'Documentation project, not infrastructure' is a limitation by the archive's measure of infrastructure; #1003's contrary grouping is composed beside #725 under P_coh; the archive's figures beyond the field (Nature Machine Intelligence 2024; 85% and 30%, D725-02) are carried at their own grade (attributed) and opened for resolution.", "considered": [ { "value": "constructs", "set_aside": "the archive does not read the Explorer or its audit as made by receiving texts" }, { "value": "revises", "set_aside": "the entity is not the archive's own; the MMRS record's correction of its scale (CAP-1-02) revises the record, not the entity" } ] }, "translation": [ { "lineage": "what it is: the explorer of an audit of training datasets", "field_claims": [ "F1", "F2", "F3", "F16", "F26", "F37" ], "received": "F1 (B1, /explorer-introduction (d2): headings): \"The Data Provenance Explorer … Explore the datasets included in our audit. The DPI Explorer tool allows you to filter for and analyze language model training datasets.\"; F2 (B1, /explorer-introduction (d2): slide caption): \"Select and filter use cases, languages, and more\"; F3 (B2, card snippet (MIT Media Lab unreachable: TLS failure)): \"Based on our extensive audits of AI data, we've built data explorers to allow practitioners to make more nuanced and informed decisions about th\"; F16 (B4, /data-provenance-explorer (p2): heading): \"Data Provenance Explorer … Discover the datasets featured in our audit! The DPI Explorer tool enables you to filter and analyze language model training datasets effortlessly.\"; F26 (B6, article body ¶4): \"The platform allows developers to track and filter thousands of datasets for legal and ethical considerations, and enables scholars and journalists to explore the composition and data lineage of popular AI datasets.\"; F37 (B7, Abstract): \"we release our entire audit, with an interactive UI, the Data Provenance Explorer, which allows practitioners to trace and filter on data provenance for the most popular open source finetuning data collections\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-01, MMRS-10, MMRS-04", "archive_claims": [ "MMRS-01", "MMRS-10", "MMRS-04" ] }, { "lineage": "the operator: the Data Provenance Initiative", "field_claims": [ "F4", "F5", "F13" ], "received": "F4 (B2, card title (MIT Media Lab unreachable)): \"Data Provenance Explorers - MIT Media Lab\"; F5 (B3, dataprovenance.org: meta description (served HTML)): \"Public measurement infrastructure for AI data, model ecosystems, and real-world AI use.\"; F13 (B3, home (Cr): 'Public-interest infrastructure'): \"DPI releases papers, datasets, and dashboards that help researchers, journalists, policymakers, and builders reason from shared evidence.\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-09, MMRS-03, MMRS-17", "archive_claims": [ "MMRS-09", "MMRS-03", "MMRS-17" ] }, { "lineage": "count: nearly 4,000 text, speech and video datasets", "field_claims": [ "F6", "F17", "F19" ], "received": "F6 (B3, home (Cr): flagship system 'Data Provenance Explorer'): \"Data Provenance Explorer … Training data … 4,000+ datasets … 608 languages across text, speech, and video\"; F17 (B4, /data-provenance-explorer (f2): 'Dataset Explorer'): \"A tool to explore and filter 4000 text, video, and speech datasets, for their license, sources, creators, and other metadata.\"; F19 (B5, /data-provenance-explorer/dataset-explorer (u2): header): \"Dataset Explorer … Explore nearly 4,000 text, video, and speech datasets, and their characteristics.\"", "receiving_text": "—", "translation": "untranslated: composed as received beside CAP-1-01, MMRS-12", "archive_claims": [ "CAP-1-01", "MMRS-12" ] }, { "lineage": "the unit: a dataset traced to its sources, licences and creators", "field_claims": [ "F7", "F21", "F25", "F43" ], "received": "F7 (B3, home (Cr): flagship system 'Data Provenance Explorer', description): \"Inspect dataset sources, licenses, creators, geographies, modalities, and derivation chains.\"; F21 (B5, /data-provenance-explorer/dataset-explorer (u2): summary table): \"Text Sources … Model Generated … Mean Input Chars … Mean Target Chars … Generated By … % Datasets Used\"; F25 (B6, article body ¶3): \"The result of this multidisciplinary initiative is the single largest audit to date of AI datasets. For the first time, these datasets include tags to the original data sources, numerous re-licensings, creators, and other data properties.\"; F43 (B9, header row of /data_summaries.csv (3,917 rows at fetch)): \"Unique Dataset Identifier,Dataset Name … Text Sources,Model Generated,Human Annotation,Derived from Datasets,Creators,Licenses … License Use (DataProvenance),License Attribution (DataProvenance),License Share Alike (DataProvenance),Modality\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-11, MMRS-16", "archive_claims": [ "MMRS-11", "MMRS-16" ] }, { "lineage": "the Web Data Explorer and Consent in Crisis", "field_claims": [ "F8", "F12", "F18", "F38", "F39", "F44", "F45" ], "received": "F8 (B3, home (Cr): flagship system 'Web Consent Explorer'): \"Web Consent Explorer … Web data … 14,000 domains … Terms, robots.txt, and crawl restrictions … Track how web consent signals are changing as AI developers compete for training data.\"; F12 (B3, home (Cr): publications, 2024): \"Consent in Crisis: The Rapid Decline of the AI Data Commons … An audit of terms, restrictions, and consent signals across the web. … NeurIPS 2024\"; F18 (B4, /data-provenance-explorer (f2): 'Web Data Explorer'): \"An explorer of the trends in data-use preferences from the websites commonly used to source AI data.\"; F38 (B8, abstract page header): \"[Submitted on 20 Jul 2024 (v1), last revised 24 Jul 2024 (this version, v2)] … Title:Consent in Crisis: The Rapid Decline of the AI Data Commons … Authors:Shayne Longpre, Robert Mahari, Ariel Lee\"; F39 (B8, Abstract): \"we conduct the first, large-scale, longitudinal audit of the consent protocols for the web domains underlying AI training corpora. Our audit of 14,000 web domains provides an expansive view of crawlable web data and how codified data use preferences are changing over time.\"; F44 (B10, /data-provenance-explorer/web-data-explorer (ed): header): \"Web Data Explorer … Shifting preferences in how websites allow their data to be used for AI\"; F45 (B10, web-data-explorer ($h): 'Trends for Robots.txt and Terms of Service'): \"We conduct a temporal analysis, from 2016 to April 2024, of the web consent signals in HEAD-C4, 2k of the largest and most critical web domains. … A web domain is considered restricted if its robots.txt disallows access to AI crawlers from OpenAI, Google, Anthropic, Common Crawl, Cohere, Meta, or the Internet Archive.\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-13, D98-02", "archive_claims": [ "MMRS-13", "D98-02" ] }, { "lineage": "the operator's other measurement systems", "field_claims": [ "F9" ], "received": "F9 (B3, home (Cr): 'Flagship systems'): \"Four public measurement systems … Together they trace AI from source data and web consent to model markets and real-world use.\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-07, MMRS-08, MMRS-14", "archive_claims": [ "MMRS-07", "MMRS-08", "MMRS-14" ] }, { "lineage": "the multimodal audit (ICLR 2025)", "field_claims": [ "F10", "F11" ], "received": "F10 (B3, home (Cr): findings strip, 'Dataset restrictions'): \"Dataset restrictions … Source content often carries stricter terms than dataset releases … Multimodal Data Provenance\"; F11 (B3, home (Cr): publications, 2025): \"Bridging the Data Provenance Gap Across Text, Speech, and Video … An audit of four thousand public text, speech, and video datasets. … ICLR 2025\"", "receiving_text": "—", "translation": "untranslated: composed as received beside no archive claim (the field's own sense)", "archive_claims": [] }, { "lineage": "reception: press and governance", "field_claims": [ "F14", "F15" ], "received": "F14 (B3, home (Cr): 'Featured coverage'): \"DPI findings have been covered by the Financial Times, The New York Times, The Washington Post, MIT Technology Review, IEEE Spectrum, and VentureBeat.\"; F15 (B3, home (Cr): 'Policy and governance relevance'): \"Our work has been cited in public AI safety and governance reporting and used to ground debates about data transparency, consent, and market concentration.\"", "receiving_text": "—", "translation": "untranslated: composed as received beside CAP-1-05, CAP-1-06, MMRS-21", "archive_claims": [ "CAP-1-05", "CAP-1-06", "MMRS-21" ] }, { "lineage": "origin: the founding audit and the launch, October 2023", "field_claims": [ "F20", "F22", "F24", "F32", "F33", "F34" ], "received": "F20 (B5, breadcrumb (Qh) and header buttons (Lh)): \"Data Provenance Explorer … Paper … Repository\"; F22 (B6, article body: standfirst): \"Researchers from MIT, Cohere for AI, and 11 other institutions launch free Data Provenance Explorer platform to track and filter audited datasets for ethical, legal, and transparency considerations\"; F24 (B6, article body ¶2): \"audit and trace nearly 2,000 of the most widely used fine-tuning datasets. Collectively, these data sets have been downloaded tens of millions of times, and are the backbone of many published NLP breakthroughs.\"; F32 (B6, Path Forward ¶2): \"This large-scale audit is the launch of a wider multi-institutional initiative, where users all over the world can contribute to the explorer\"; F33 (B7, abstract page header): \"[Submitted on 25 Oct 2023 (v1), last revised 4 Nov 2023 (this version, v3)] … Title:The Data Provenance Initiative: A Large Scale Audit of Dataset Licensing & Attribution in AI … Authors:Shayne Longpre, Robert Mahari, Anthony Chen, Naana Obeng-Marnu\"; F34 (B7, Abstract): \"we convene a multi-disciplinary effort between legal and machine learning experts to systematically audit and trace 1800+ text datasets. We develop tools and standards to trace the lineage of these datasets, from their source, creators, series of license conditions, properties, and subsequent use.\"", "receiving_text": "—", "translation": "untranslated: composed as received beside D1714-02, D1714-03", "archive_claims": [ "D1714-02", "D1714-03" ] }, { "lineage": "data provenance defined", "field_claims": [ "F23" ], "received": "F23 (B6, article body ¶1): \"Inconsistent tracking of the origins of a piece of data, which we refer to as data provenance, has introduced a range of legal and ethical challenges.\"", "receiving_text": "—", "translation": "untranslated: composed as received beside O725-03, O725-05, O725-07", "archive_claims": [ "O725-03", "O725-05", "O725-07" ] }, { "lineage": "audit findings: licence omission and misattribution", "field_claims": [ "F27", "F28", "F36" ], "received": "F27 (B6, Addressing the Data Transparency Crisis ¶1): \"Crowdsourced aggregators like GitHub, Papers with Code, and many of the open source LLMs trained from data on these aggregators, have an extremely high proportion of missing data licenses (“Unspecified”), ranging from 72 to 83 percent. This compares to 30 percent missing licenses with our own annotation protocol\"; F28 (B6, Addressing the Data Transparency Crisis ¶2): \"Crowdsourced aggregators listed licenses too permissively in up to 29 percent of the various cases audited. Incorrect or distorted licensing leads to cascading misattribution\"; F36 (B7, Abstract): \"We also observe frequent miscategorization of licenses on widely used dataset hosting sites, with license omission of 70%+ and error rates of 50%+. This points to a crisis in misattribution and informed use of the most popular datasets driving many recent breakthroughs.\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-02, MMRS-05, D725-02, D1714-01", "archive_claims": [ "MMRS-02", "MMRS-05", "D725-02", "D1714-01" ] }, { "lineage": "audit findings: the commercial divide and the Western skew", "field_claims": [ "F29", "F30", "F35" ], "received": "F29 (B6, Understanding Systematic Differences ¶2): \"There is a sharp and widening divide between data licensed as commercially open versus closed.\"; F30 (B6, Understanding Systematic Differences ¶4): \"We find a stark Western-centric skew in representation across datasets.\"; F35 (B7, Abstract): \"Our landscape analysis highlights the sharp divides in composition and focus of commercially open vs closed datasets, with closed datasets monopolizing important categories: lower resource languages, more creative tasks, richer topic variety, newer and more synthetic training data.\"", "receiving_text": "—", "translation": "untranslated: composed as received beside no archive claim (the field's own sense)", "archive_claims": [] }, { "lineage": "stated limits: what tooling cannot resolve", "field_claims": [ "F31" ], "received": "F31 (B6, Open Legal Ambiguities ¶1): \"major legal ambiguities remain in data licensing that the authors acknowledge cannot be resolved by tooling alone.\"", "receiving_text": "—", "translation": "untranslated: composed as received beside D725-04, D725-08", "archive_claims": [ "D725-04", "D725-08" ] }, { "lineage": "consent findings: the decline of the AI data commons", "field_claims": [ "F40", "F41", "F42", "F46" ], "received": "F40 (B8, Abstract): \"in a single year (2023-2024) there has been a rapid crescendo of data restrictions from web sources, rendering ~5%+ of all tokens in C4, or 28%+ of the most actively maintained, critical sources in C4, fully restricted from use. For Terms of Service crawling restrictions, a full 45% of C4 is now restricted.\"; F41 (B8, Abstract): \"We diagnose these as symptoms of ineffective web protocols, not designed to cope with the widespread re-purposing of the internet for AI.\"; F42 (B8, Abstract): \"If respected or enforced, these restrictions are rapidly biasing the diversity, freshness, and scaling laws for general-purpose AI systems.\"; F46 (B10, web-data-explorer ($h): 'Trends for Robots.txt and Terms of Service'): \"In all cases there are significant recent spikes in restrictions.\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-06", "archive_claims": [ "MMRS-06" ] }, { "lineage": "reception: the encyclopedia", "field_claims": [ "F47", "F48" ], "received": "F47 (B11, Wikipedia: title Data_Provenance_Explorer): \"raw wikitext (action=raw, redirect=no): HTTP 404 … Wikipedia does not have an article with this exact name. … There were no results matching the query.\"; F48 (B12, Wikipedia: title Data_Provenance_Initiative): \"raw wikitext (action=raw, redirect=no): HTTP 404 … Wikipedia does not have an article with this exact name. … There were no results matching the query.\"", "receiving_text": "—", "translation": "untranslated: composed as received beside CAP-1-09, CAP-1-08, CAP-1-03", "archive_claims": [ "CAP-1-09", "CAP-1-08", "CAP-1-03" ] } ], "receiving_texts": "equal-terms case: every field claim composed as received, beside the archive claims it stands with" }, "measure": { "rule": "v0.9 §4.10: KO_B sentences surviving in the KO at similarity ≥ 0.95, with positions (computed as apply_v09.py does: difflib.SequenceMatcher ratio, the best KO sentence for each KO_B sentence)", "v0.9": { "surviving": 0, "of": 12, "positions": [ null, null, null, null, null, null, null, null, null, null, null, null ] }, "opening_run": { "rule": "v0.9 §5.10 (equal-terms case): the field-only KO's sentences as the KO's opening run in their own order", "breach": false }, "blocks": { "rule": "v0.9 §4.1 (equal-terms and mixed cases): the KO does not open with the field as a block or close with the archive as one; sentences typed by the ledgers of the claims they cite (F field-only, A archive-only, M both)", "sequence": "MFFMFMFAFFMAMFFMFMAMAAMAMAMAM", "field_only_head": 0, "archive_only_tail": 0, "breach": false }, "max_similarity": [ 0.54, 0.29, 0.284, 0.367, 0.298, 0.259, 0.482, 0.277, 0.274, 0.297, 0.227, 0.364 ], "ko_sentence_1_max_similarity_to_KO_B": 0.29 }, "homonyms": { "names": [], "basis": "The MMRS census row carries no disambiguation field (nor negative_tags or battery); no card is a homonym. 'Data Provenance Explorers' (MIT Media Lab's title, plural) names the same tool. 'DPI', the Initiative's own short form, is in the archive also #261's distinction preservation index (a metric of the specification, §6.4) and is not an alias. The general concept 'data provenance' (the tracking of the origins of data, F23; in computer science W3C PROV, named by #725) is a sense of the term, not of the entity.", "distinct_entities": [ "Data Provenance Initiative (the operator)", "Data-Provenance-Collection (the operator's GitHub repository)", "Open Model Evolution and the AI Observatory (the Initiative's other measurement systems)", "papers-with-code (named in B6 as a crowdsourced aggregator the audit measured)", "C2PA / Content Credentials and W3C PROV (the other provenance frameworks #725 sets beside it)" ], "substitutes": { "T": [] }, "register": { "address_1": { "slug": "data-provenance-explorer-aio-20261029", "obs_id": "OBS-dc0adc018f10" } } }, "notes": [ "One address, Google AI Overview (expanded), signed out, incognito, 2026-10-29: data-provenance-explorer-aio-20261029 (OBS-dc0adc018f10, ADDR-6e5a719fee75, AXN:0AAC.CAPTURE.🕔🎨🟢●🌺☉), six cards, 110 words in the answer body. A NEW address; the MMRS record landed the same night and no card cites it (CAP-1-08).", "Wikipedia: no article for the entity or its operator. The titles Data_Provenance_Explorer and Data_Provenance_Initiative return HTTP 404 for raw wikitext (redirect=no) and for the rendered page, neither is a redirect, and the rendered Special:Search for each quoted phrase in article space reports no results (B11, B12; the search API answered 429 and was not used). In a probe not saved to the intake, the general term 'Data provenance' redirects to 'Data lineage#Data_provenance' (the concept, not the entity; not ledgered).", "Fetching: every reachable page saved verbatim with curl on 2026-10-29 between 05:11 and 05:12 UTC, raw and text with sha256 in field.json. The Initiative's site is a single-page application: every path serves the same 1,601-byte shell (title 'Data Provenance Initiative', meta description, empty root). Its text was taken from the script bundle the shell loads (/static/js/main.af19d7fc.js, 2,800,478 bytes, saved whole), by route component (the router maps / to Cr, /explorer-introduction to d2, /data-provenance-explorer to p2, /data-provenance-explorer/dataset-explorer to u2, /data-provenance-explorer/web-data-explorer to ed), extracting the component's string literals in source order; the Dataset Explorer's data is /data_summaries.csv, which the bundle fetches (3,917 rows: 3,717 text, 105 video, 95 speech; 308 collections), saved gzip-compressed with the served file's sha256 recorded. Card URLs: B1, B3, B4, B5 identified by matching each card's snippet to a route component's text (the cards' own URLs are not in the paste); B2 by WebSearch on title and site; B6 from the Initiative's press list in the bundle, confirmed by the page title. Cohere's article body is carried in the page's Next.js flight data, extracted and appended to B6.txt under a marked header.", "Unreachable: B2, MIT Media Lab's project page (media.mit.edu/projects/data-provenance-for-ai/tools): three fetches (the brief's command; with the session's CA bundle; the www-prod host) failed TLS verification (curl 60, self-signed certificate in chain) before any HTTP response. A request for an archived copy was refused by the workspace and not pursued. Its title and snippet stand (F3, F4).", "The archive subset: the string pass (non_traverse.py) returned 35 D, 5 R and 27 O candidates in 10 deposits; reading admitted 10 deposits, 6 carrying claims (#725, #98, #1714, #1003, #95, #623) and 4 grouped as instances under them by S4 (#1190 under #725, #121 and #1194 under #1003, #1188 under #95). The hop (73 candidates in 72 deposits) admitted none. Two string-pass defects, reported, not fixed: (1) sentences are cut at 900 characters, so an alias late in a long markdown table row is missed (#98's 'Provenance schema … Implemented (in CHA)' row, R98-04; the specification's F16 row in #1664, #1665, #1671, #1690); (2) the O-class pass did not return #95/#1188, whose class sentence ('training-data provenance records') carries no category from the script's vocabulary. Both were found by reading the 17 texts the task names.", "Defects in the MMRS record (reported, not fixed): (a) self_description's evidence is arxiv.org/abs/2310.16787 (basis operator), but the phrase 'Public measurement infrastructure for AI data, model ecosystems, and real-world AI use' is the dataprovenance.org meta description (B3) and is not on the founding paper's abstract page (B7); (b) the note's 'NeurIPS 2024 Datasets and Benchmarks' is stated by no source read (the home page has 'NeurIPS 2024', the arXiv page no venue; its evidence is arxiv.org/abs/2407.14933); (c) unit and count carry basis 'operator' with the capture as src: the text is the operator's, as seen on the composition layer's cards, which the sources entry states; (d) 'unchecked: the explorers' pages directly (they render by script)': their text is readable in the site's script bundle and the Dataset Explorer's data in /data_summaries.csv without rendering (this row's B1, B3–B5, B9, B10); (e) the census row has no disambiguation, negative_tags or battery fields, which the brief expects of a record; (f) outside this entity: articles.py's ARTICLES['oecd-ai-incidents-monitor'] holds the CodeSOTA article (its kernel begins 'CodeSOTA, by Kacper Wikiel'), so the OECD monitor's record carries another entity's text.", "Observations on the capture (reported, not fixed): findings[0] says the Initiative's pages 'render nothing to this record's fetcher beyond a title' — true of the served HTML, which also carries the meta description; the page text is in the bundle. The founding audit's scale is held two ways in the field (B7 '1800+ text datasets'; B6 'nearly 2,000 … fine-tuning datasets'), and the record and the capture follow the paper's.", "The archive's own figures beyond the field: #725 and #1190 give the founding audit's venue as Nature Machine Intelligence 2024 and its findings as '85% of licenses request attribution and 30% include share-alike clauses' (D725-02); no field source read states either. Carried at 'attributed' and opened.", "Composer's reading of the case: equal (ρ = extends), from the MMRS record and #725's own statement of relation ('Extension and completion, not substitution or critique'); see ontology.case.", "Shapes for the page builder: objects.T is one dict with one transcript; field is flat with field_by_origin beside; ontology.composition 'equal'; E_A.defining_papers integers, defining_records strings; CAP- and MMRS- claims carry 'dep' as a string, 'surface' and 'record'; rails give 'earliest' as a field source id, a deposit number, 'MMRS' or 'capture:' (as the two finished rows do). build_non.py, non_axn.py and panel.json untouched." ], "composition_v09": { "source": "this row (composed directly; no rebuild/negative-of-the-negative/v09/data-provenance-explorer.json)", "composer": "Claude (Opus 5.5), subagent, 2026-10-29", "defining_material_read": [ { "what": "MMRS record, census row", "file": "/home/claude/fleet/machinemediation-org/data/meta-aggregator/census.json (rows[slug == 'data-provenance-explorer'], schema_note)", "read": "whole" }, { "what": "MMRS record, article", "file": "/home/claude/fleet/machinemediation-org/scripts/meta_aggregator/articles.py (ARTICLES['data-provenance-explorer'])", "read": "whole" }, { "what": "MMRS record, rendered", "file": "/home/claude/fleet/machinemediation-org/aggregators/data-provenance-explorer/index.html", "read": "text, for agreement with the sources" }, { "what": "capture record", "file": "data/EA-WG-CAPTURES-01.json (data-provenance-explorer-aio-20261029)", "read": "transcript, transcript_raw, cite_list, d, reading, findings, analysis, q_kind, per_note" }, { "dep": 725, "file": "data/texts/AXN-0273-text.md", "read": "the dimensions, §1.2, §3.1–3.5 and the closing, around every passage naming the Initiative" }, { "dep": 98, "file": "data/texts/AXN-0271-text.md", "read": "§D and the layer and component tables" }, { "dep": 1714, "file": "data/texts/AXN-0A7F-text.md", "read": "§4.1, §6.6, §9.2 and the revision notes naming the Initiative" }, { "dep": 95, "file": "data/texts/AXN-026A-text.md", "read": "the typology of source-dependent claims" }, { "dep": 1003, "file": "data/texts/AXN-03F7-text.md", "read": "the paragraph naming the Initiative" }, { "dep": 623, "file": "data/texts/AXN-01DA-text.md", "read": "§8.3" }, { "what": "the specification", "file": "data/texts/AXN-0700-text.md", "read": "§0–§11 whole; Appendix A by its headings" } ], "cuts": [], "notes": "equal case; KO 29 sentences, all 48 field and 49 archive claims cited; P body 344 words; KO_B 12 sentences, none surviving at ≥ 0.95" }, "field_by_origin": { "cards": [ "B1", "B2", "B3", "B4", "B5", "B6" ], "own_pages": [ "B7", "B8", "B9", "B10" ], "wikipedia": [ "B11", "B12" ], "unreachable": [ "B2" ], "title_only": [ "B2" ], "script_rendered": [ "B1", "B3", "B4", "B5", "B10" ], "wikipedia_state": "no article: 'Data Provenance Explorer' and 'Data Provenance Initiative' are both absent (raw wikitext HTTP 404, rendered page HTTP 404 'Wikipedia does not have an article with this exact name'), no redirect, and a full-text search of article space for each quoted phrase returns 'There were no results matching the query.' (2026-10-29)", "sources": [ { "id": "B1", "origin": "card", "role": "Overview card 1", "card": "Data Provenance Initiative — Data Provenance Explorer", "url": "https://www.dataprovenance.org/explorer-introduction", "kind": "Overview card (own page)", "surfaced_by_T": [ { "slug": "data-provenance-explorer-aio-20261029", "card": 1 } ], "files": { "B1.html": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B1.html", "sha256": "bf609f385257c3a179dc564d68d8ec46b2f096a34f61918cc7d750b83f6d7e2f" }, "B1.txt": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B1.txt", "sha256": "2af1b48c11922848d4a50b1d5254e79d0900e539ab52d001e210628b11e5d7e1" }, "site-main.af19d7fc.js": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/site-main.af19d7fc.js", "sha256": "91de539c0ec5317faec490142648792d586b0ee21e82a88f5d7feba90a0898e9" } }, "fetched": { "at": "2026-10-29T05:12:02Z", "http": 200, "url": "https://www.dataprovenance.org/explorer-introduction", "raw": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B1.html", "raw_sha256": "bf609f385257c3a179dc564d68d8ec46b2f096a34f61918cc7d750b83f6d7e2f", "text": [ "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B1.txt" ], "bundle": { "url": "https://www.dataprovenance.org/static/js/main.af19d7fc.js", "at": "2026-10-29T05:12:01Z", "http": 200, "sha256": "91de539c0ec5317faec490142648792d586b0ee21e82a88f5d7feba90a0898e9" } }, "card_title": "Data Provenance Explorer", "card_snippet": "Data Provenance Initiative The Data Provenance Explorer Explore the datasets included in our audit. The DPI Explorer tool allows you to filter for and analyze l...", "card_note": null, "names_operator": "yes ('DPI', 'Data Provenance Initiative' in the saved text)", "names_archive": "no (none of the Crimson Hexagonal Archive, Alexanarch, Lee Sharks, machinemediation, EA-WG-CAPTURES-01 in the saved text, card title and snippet)", "note": "URL identified by the card's text: the snippet's 'Explore the datasets included in our audit. The DPI Explorer tool allows you to filter for and analyze l…' is the heading of the /explorer-introduction route's component (d2) and occurs nowhere else in the bundle; the card's own URL is not in the paste. The served HTML is the site's script shell (title, meta description, an empty root; the same 1,601 bytes at every path, sha256 bf609f38…); the page text is the string literals of the route's component in the site's script bundle /static/js/main.af19d7fc.js, fetched verbatim and saved as site-main.af19d7fc.js, extracted in source order by a tokenizing pass (component and offsets named in the .txt)." }, { "id": "B2", "origin": "card", "role": "Overview card 2", "card": "MIT Media Lab — Data Provenance Explorers - MIT Media Lab", "url": "https://www.media.mit.edu/projects/data-provenance-for-ai/tools", "kind": "Overview card (unreachable)", "surfaced_by_T": [ { "slug": "data-provenance-explorer-aio-20261029", "card": 2 } ], "files": {}, "fetched": { "unreachable": true, "tries": [ { "at": "2026-10-29T05:12:03Z", "url": "https://www.media.mit.edu/projects/data-provenance-for-ai/tools", "http": 0, "error": "curl: (60) SSL certificate problem: self-signed certificate in certificate chain" }, { "at": "2026-10-29T05:12:03Z", "url": "https://www.media.mit.edu/projects/data-provenance-for-ai/tools", "http": 0, "error": "curl: (60) SSL certificate problem: self-signed certificate in certificate chain" }, { "at": "2026-10-29T05:12:04Z", "url": "https://www-prod.media.mit.edu/projects/data-provenance-for-ai/tools", "http": 0, "error": "curl: (60) SSL certificate problem: self-signed certificate in certificate chain" } ], "note": "URL found by WebSearch on the card's title and site (media.mit.edu and www-prod.media.mit.edu, /projects/data-provenance-for-ai/tools). Three fetches, 2026-10-29: curl with the brief's command; the same with --cacert /root/.ccr/ca-bundle.crt; the www-prod host. Each failed TLS verification (curl error 60, self-signed certificate in chain) before any HTTP response. A request for an archived copy was refused by the workspace and not pursued. The card's title and snippet stand as its only text." }, "card_title": "Data Provenance Explorers - MIT Media Lab", "card_snippet": "Share this tool Based on our extensive audits of AI data, we've built data explorers to allow practitioners to make more nuanced and informed decisions about th...", "card_note": "unreachable: snippet only", "names_operator": "no (the title and snippet name neither the Initiative nor its members)", "names_archive": "no (title and snippet)", "note": "MIT Media Lab's project page for the Initiative's explorers (the plural in its title). Its claims (F3, F4) are quoted from the card's title and snippet." }, { "id": "B3", "origin": "card", "role": "Overview card 3; also the own page the MMRS record cites (dataprovenance.org)", "card": "Data Provenance Initiative — Data Provenance Initiative", "url": "https://www.dataprovenance.org/", "kind": "Overview card (own page)", "surfaced_by_T": [ { "slug": "data-provenance-explorer-aio-20261029", "card": 3 } ], "files": { "B3.html": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B3.html", "sha256": "bf609f385257c3a179dc564d68d8ec46b2f096a34f61918cc7d750b83f6d7e2f" }, "B3.txt": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B3.txt", "sha256": "d1f28f3a590faab9ab608d7781fd06b1bacce7d2901e6f65ba25de0949d620dc" }, "site-main.af19d7fc.js": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/site-main.af19d7fc.js", "sha256": "91de539c0ec5317faec490142648792d586b0ee21e82a88f5d7feba90a0898e9" } }, "fetched": { "at": "2026-10-29T05:12:02Z", "http": 200, "url": "https://www.dataprovenance.org/", "raw": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B3.html", "raw_sha256": "bf609f385257c3a179dc564d68d8ec46b2f096a34f61918cc7d750b83f6d7e2f", "text": [ "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B3.txt" ], "bundle": { "url": "https://www.dataprovenance.org/static/js/main.af19d7fc.js", "at": "2026-10-29T05:12:01Z", "http": 200, "sha256": "91de539c0ec5317faec490142648792d586b0ee21e82a88f5d7feba90a0898e9" } }, "card_title": "Data Provenance Initiative", "card_snippet": "Data Provenance Explorer. 4,000+ datasets. Inspect dataset sources, licenses, creators, geographies, modalities, and derivation chains. Explore datasets · Web d...", "card_note": null, "names_operator": "yes ('DPI', 'Data Provenance Initiative' in the saved text)", "names_archive": "no (none of the Crimson Hexagonal Archive, Alexanarch, Lee Sharks, machinemediation, EA-WG-CAPTURES-01 in the saved text, card title and snippet)", "note": "the card's snippet ('Data Provenance Explorer. 4,000+ datasets. Inspect dataset sources…') is the home page's flagship-systems card (Cr, constant xr). The served HTML is the site's script shell (title, meta description, an empty root; the same 1,601 bytes at every path, sha256 bf609f38…); the page text is the string literals of the route's component in the site's script bundle /static/js/main.af19d7fc.js, fetched verbatim and saved as site-main.af19d7fc.js, extracted in source order by a tokenizing pass (component and offsets named in the .txt)." }, { "id": "B4", "origin": "card", "role": "Overview card 4", "card": "Data Provenance Initiative — Data Provenance Explorer", "url": "https://www.dataprovenance.org/data-provenance-explorer", "kind": "Overview card (own page)", "surfaced_by_T": [ { "slug": "data-provenance-explorer-aio-20261029", "card": 4 } ], "files": { "B4.html": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B4.html", "sha256": "bf609f385257c3a179dc564d68d8ec46b2f096a34f61918cc7d750b83f6d7e2f" }, "B4.txt": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B4.txt", "sha256": "ca844fa5c36fd62d903d620f34850819565b92c92fed8b10bf23b85ad4a50025" }, "site-main.af19d7fc.js": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/site-main.af19d7fc.js", "sha256": "91de539c0ec5317faec490142648792d586b0ee21e82a88f5d7feba90a0898e9" } }, "fetched": { "at": "2026-10-29T05:12:02Z", "http": 200, "url": "https://www.dataprovenance.org/data-provenance-explorer", "raw": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B4.html", "raw_sha256": "bf609f385257c3a179dc564d68d8ec46b2f096a34f61918cc7d750b83f6d7e2f", "text": [ "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B4.txt" ], "bundle": { "url": "https://www.dataprovenance.org/static/js/main.af19d7fc.js", "at": "2026-10-29T05:12:01Z", "http": 200, "sha256": "91de539c0ec5317faec490142648792d586b0ee21e82a88f5d7feba90a0898e9" } }, "card_title": "Data Provenance Explorer", "card_snippet": "Discover the datasets featured in our audit! The DPI Explorer tool enables you to filter and analyze language model training datasets effortlessly. Dataset Expl...", "card_note": null, "names_operator": "yes ('DPI', 'Data Provenance Initiative' in the saved text)", "names_archive": "no (none of the Crimson Hexagonal Archive, Alexanarch, Lee Sharks, machinemediation, EA-WG-CAPTURES-01 in the saved text, card title and snippet)", "note": "the snippet ('Discover the datasets featured in our audit!…Dataset Expl…') is the /data-provenance-explorer route's heading and its first tile (p2, f2). The served HTML is the site's script shell (title, meta description, an empty root; the same 1,601 bytes at every path, sha256 bf609f38…); the page text is the string literals of the route's component in the site's script bundle /static/js/main.af19d7fc.js, fetched verbatim and saved as site-main.af19d7fc.js, extracted in source order by a tokenizing pass (component and offsets named in the .txt)." }, { "id": "B5", "origin": "card", "role": "Overview card 5", "card": "Data Provenance Initiative — Dataset Explorer - Data Provenance Initiative", "url": "https://www.dataprovenance.org/data-provenance-explorer/dataset-explorer", "kind": "Overview card (own page)", "surfaced_by_T": [ { "slug": "data-provenance-explorer-aio-20261029", "card": 5 } ], "files": { "B5.html": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B5.html", "sha256": "bf609f385257c3a179dc564d68d8ec46b2f096a34f61918cc7d750b83f6d7e2f" }, "B5.txt": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B5.txt", "sha256": "fb49cb6607f887ec58400e261da917030083586ce6319537d7cbae209d960c20" }, "site-main.af19d7fc.js": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/site-main.af19d7fc.js", "sha256": "91de539c0ec5317faec490142648792d586b0ee21e82a88f5d7feba90a0898e9" } }, "fetched": { "at": "2026-10-29T05:12:03Z", "http": 200, "url": "https://www.dataprovenance.org/data-provenance-explorer/dataset-explorer", "raw": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B5.html", "raw_sha256": "bf609f385257c3a179dc564d68d8ec46b2f096a34f61918cc7d750b83f6d7e2f", "text": [ "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B5.txt" ], "bundle": { "url": "https://www.dataprovenance.org/static/js/main.af19d7fc.js", "at": "2026-10-29T05:12:01Z", "http": 200, "sha256": "91de539c0ec5317faec490142648792d586b0ee21e82a88f5d7feba90a0898e9" } }, "card_title": "Dataset Explorer - Data Provenance Initiative", "card_snippet": "Data Provenance Explorer. > Dataset Explorer. Dataset Explorer. Explore nearly 4,000 text, video, and speech datasets, and their characteristics. Paper Reposito...", "card_note": null, "names_operator": "yes ('Data Provenance Initiative' in the saved text)", "names_archive": "no (none of the Crimson Hexagonal Archive, Alexanarch, Lee Sharks, machinemediation, EA-WG-CAPTURES-01 in the saved text, card title and snippet)", "note": "the snippet's 'Data Provenance Explorer. > Dataset Explorer.' is the breadcrumb (Qh) and 'Paper Reposito…' the header's two buttons (Lh, 'Paper' and 'Repository'); the buttons link the founding paper (https://arxiv.org/pdf/2310.16787, in the component) and the Data-Provenance-Collection repository. The served HTML is the site's script shell (title, meta description, an empty root; the same 1,601 bytes at every path, sha256 bf609f38…); the page text is the string literals of the route's component in the site's script bundle /static/js/main.af19d7fc.js, fetched verbatim and saved as site-main.af19d7fc.js, extracted in source order by a tokenizing pass (component and offsets named in the .txt)." }, { "id": "B6", "origin": "card", "role": "Overview card 6", "card": "Cohere — Data Provenance Explorer Launches to Tackle Data Transparency Crisis", "url": "https://cohere.com/blog/data-provenance-explorer-launches-tackle-data-transparency-crisis", "kind": "Overview card (press post by the operators)", "surfaced_by_T": [ { "slug": "data-provenance-explorer-aio-20261029", "card": 6 } ], "files": { "B6.html": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B6.html", "sha256": "aa465dcc01c69325c46d7abecf7b3c3f60d80cec9522c1f4b54961cb9752dc80" }, "B6.txt": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B6.txt", "sha256": "77dc370bbe2dfad003b8ff5176c86dcfdba0c7f978851c91d3cddf75b2d8bb5e" } }, "fetched": { "at": "2026-10-29T05:12:04Z", "http": 200, "url": "https://cohere.com/blog/data-provenance-explorer-launches-tackle-data-transparency-crisis", "raw": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B6.html", "raw_sha256": "aa465dcc01c69325c46d7abecf7b3c3f60d80cec9522c1f4b54961cb9752dc80", "text": [ "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B6.txt" ] }, "card_title": "Data Provenance Explorer Launches to Tackle Data Transparency Crisis", "card_snippet": "By Shayne Longpre and Sara Hooker Collectively, these data sets have been downloaded tens of millions of times, and are the backbone of many published NLP break...", "card_note": null, "names_operator": "yes ('Data Provenance Initiative', 'Longpre' in the saved text)", "names_archive": "no (none of the Crimson Hexagonal Archive, Alexanarch, Lee Sharks, machinemediation, EA-WG-CAPTURES-01 in the saved text, card title and snippet)", "note": "URL from the Initiative's own press list in its script bundle (cohere.com/blog/data-provenance-explorer-launches-tackle-data-transparency-crisis), confirmed by the page title. Dated 25 October 2023; by Shayne Longpre and Sara Hooker. The page's served text holds only navigation; the article body is in its Next.js flight data (record 29:T20fe, HTML), extracted and appended to B6.txt under a marked header." }, { "id": "B7", "origin": "own_page", "role": "own page the MMRS record cites (the operators' paper)", "card": "arXiv — [2310.16787] The Data Provenance Initiative: A Large Scale Audit of Dataset Licensing & Attribution in AI", "url": "https://arxiv.org/abs/2310.16787", "kind": "own page (abstract page)", "surfaced_by_T": "no (not surfaced by T; admitted beside the cards by the operator's ruling of 2026-10-29)", "files": { "B7.html": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B7.html", "sha256": "d01552052066b391aeed8b50adde61897cb194ea14727553cbf65971af51bb53" }, "B7.txt": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B7.txt", "sha256": "cf127e18a98a4ee768753591e9e8a09c79c6b92f8ead98a6a1dd45f62a036504" } }, "fetched": { "at": "2026-10-29T05:12:04Z", "http": 200, "url": "https://arxiv.org/abs/2310.16787", "raw": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B7.html", "raw_sha256": "d01552052066b391aeed8b50adde61897cb194ea14727553cbf65971af51bb53", "text": [ "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B7.txt" ] }, "names_operator": "yes ('Data Provenance Initiative', 'Longpre', 'Mahari' in the saved text)", "names_archive": "no (none of the Crimson Hexagonal Archive, Alexanarch, Lee Sharks, machinemediation, EA-WG-CAPTURES-01 in the saved text, card title and snippet)", "note": "the founding paper's abstract page; cited by the MMRS record (sources: arxiv.org/abs/2310.16787, for 'audit'; evidence for operator, self_description and records_attribution)" }, { "id": "B8", "origin": "own_page", "role": "own page the MMRS record cites (the operators' paper)", "card": "arXiv — [2407.14933] Consent in Crisis: The Rapid Decline of the AI Data Commons", "url": "https://arxiv.org/abs/2407.14933", "kind": "own page (abstract page)", "surfaced_by_T": "no (not surfaced by T; admitted beside the cards by the operator's ruling of 2026-10-29)", "files": { "B8.html": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B8.html", "sha256": "8cedde3659e5ed29df809dfd194af1b64f4bcf5c215933d3b1741f1ad120c98b" }, "B8.txt": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B8.txt", "sha256": "d90c4cbbca83904750a946084c4adb455164500d67431e02f7c8bd90868c07fd" } }, "fetched": { "at": "2026-10-29T05:12:04Z", "http": 200, "url": "https://arxiv.org/abs/2407.14933", "raw": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B8.html", "raw_sha256": "8cedde3659e5ed29df809dfd194af1b64f4bcf5c215933d3b1741f1ad120c98b", "text": [ "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B8.txt" ] }, "names_operator": "yes ('Longpre', 'Mahari' in the saved text)", "names_archive": "no (none of the Crimson Hexagonal Archive, Alexanarch, Lee Sharks, machinemediation, EA-WG-CAPTURES-01 in the saved text, card title and snippet)", "note": "the second audit's abstract page; cited by the MMRS record (sources: arxiv.org/abs/2407.14933, for 'Consent in Crisis audit'; evidence for note)" }, { "id": "B9", "origin": "own_page", "role": "the data the Dataset Explorer loads (the JSON/CSV the page fetches)", "card": "Data Provenance Initiative — /data_summaries.csv", "url": "https://www.dataprovenance.org/data_summaries.csv", "kind": "own page (data file)", "surfaced_by_T": "no (not surfaced by T; found in the script bundle, which fetches it for the Dataset Explorer)", "files": { "B9.csv.gz": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B9.csv.gz", "sha256": "e1eb7bf01f96315f5c7a6261e66afbf12cbc66328d3904dd0adf153a2f22c9e7" }, "B9.txt": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B9.txt", "sha256": "65ff8518cd883799afa5d68f7502269a2f56ca5b24dfe89497e49ef6b7f4c10d" } }, "fetched": { "at": "2026-10-29T05:12:04Z", "http": 200, "url": "https://www.dataprovenance.org/data_summaries.csv", "content_type": "text/csv; charset=utf-8", "raw": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B9.csv.gz", "raw_sha256_uncompressed": "ca7049b3f307ef6cd3bdb409e7184e054c42816d1551cdfe918a6f6b081658ad", "raw_sha256": "e1eb7bf01f96315f5c7a6261e66afbf12cbc66328d3904dd0adf153a2f22c9e7", "text": [ "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B9.txt" ], "rows": 3917, "by_modality": { "Text": 3717, "Speech": 95, "Video": 105 }, "by_license_use": { "unspecified": 701, "non-commercial": 873, "commercial": 2116, "academic-only": 227 }, "collections": 308, "note": "13,152,102 bytes as served, saved gzip-compressed (raw_sha256_uncompressed is the served file's); rows counted with Python's csv reader; B9.txt holds the header row verbatim" }, "names_operator": "yes (the header's '(DataProvenance)' columns are the Initiative's own licence readings)", "names_archive": "no", "note": "The bundle also loads /constants/*.json (licence classes, creator, domain, task and model groups) and /html_constants/*.json (country and language codes), not fetched: lookup tables for the filters." }, { "id": "B10", "origin": "own_page", "role": "the Initiative's explorer page for web data (task: its explorer pages)", "card": "Data Provenance Initiative — Web Data Explorer", "url": "https://www.dataprovenance.org/data-provenance-explorer/web-data-explorer", "kind": "own page (explorer)", "surfaced_by_T": "no (not surfaced by T, whose web-data bullet has no card of its own; admitted beside the cards by the operator's ruling of 2026-10-29)", "files": { "B10.html": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B10.html", "sha256": "bf609f385257c3a179dc564d68d8ec46b2f096a34f61918cc7d750b83f6d7e2f" }, "B10.txt": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B10.txt", "sha256": "ab370b14791658a228880d53c69187198ba2a6cf80a2f8daea5a251ad155679a" }, "site-main.af19d7fc.js": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/site-main.af19d7fc.js", "sha256": "91de539c0ec5317faec490142648792d586b0ee21e82a88f5d7feba90a0898e9" } }, "fetched": { "at": "2026-10-29T05:12:03Z", "http": 200, "url": "https://www.dataprovenance.org/data-provenance-explorer/web-data-explorer", "raw": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B10.html", "raw_sha256": "bf609f385257c3a179dc564d68d8ec46b2f096a34f61918cc7d750b83f6d7e2f", "text": [ "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B10.txt" ], "bundle": { "url": "https://www.dataprovenance.org/static/js/main.af19d7fc.js", "sha256": "91de539c0ec5317faec490142648792d586b0ee21e82a88f5d7feba90a0898e9" } }, "names_operator": "yes ('Data Provenance Initiative' in the saved text)", "names_archive": "no (none of the Crimson Hexagonal Archive, Alexanarch, Lee Sharks, machinemediation, EA-WG-CAPTURES-01 in the saved text, card title and snippet)", "note": "the route's component ($h, ed) with the shared breadcrumb (Qh) and header buttons (Lh); its paper link is /Consent_in_Crisis.pdf. The served HTML is the site's script shell (title, meta description, an empty root; the same 1,601 bytes at every path, sha256 bf609f38…); the page text is the string literals of the route's component in the site's script bundle /static/js/main.af19d7fc.js, fetched verbatim and saved as site-main.af19d7fc.js, extracted in source order by a tokenizing pass (component and offsets named in the .txt)." }, { "id": "B11", "origin": "wikipedia", "role": "the entity's Wikipedia title", "card": "Wikipedia — Data Provenance Explorer (no article)", "url": "https://en.wikipedia.org/wiki/Data_Provenance_Explorer", "kind": "Wikipedia (no article)", "surfaced_by_T": "no (not surfaced by T; checked by the operator's ruling of 2026-10-29)", "files": { "B11.wiki": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B11.wiki", "sha256": "ae88bc8014d236e3d1d5d71f37597fcd135734056aa3c713db66d4aa0999c7a0" }, "B11.html": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B11.html", "sha256": "136eb91afa8a5c7f30f34144a07ff5e379bb00c32b6cfbfa6c8953894f865c56" }, "B11-search.html": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B11-search.html", "sha256": "d510d4b8db965748ad8f7fb7fe5406329f363093ef85122c6688c98e0f944658" }, "B11.txt": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B11.txt", "sha256": "4f6b81e12554dfb5c6af75dd107702f06a9aa8ec016ce056c21a02f992a5bc43" } }, "fetched": { "raw": { "url": "https://en.wikipedia.org/w/index.php?title=Data_Provenance_Explorer&action=raw&redirect=no", "at": "2026-10-29T05:12:05Z", "http": 404, "sha256": "ae88bc8014d236e3d1d5d71f37597fcd135734056aa3c713db66d4aa0999c7a0" }, "rendered": { "url": "https://en.wikipedia.org/wiki/Data_Provenance_Explorer", "at": "2026-10-29T05:12:06Z", "http": 404, "sha256": "136eb91afa8a5c7f30f34144a07ff5e379bb00c32b6cfbfa6c8953894f865c56" }, "search": { "url": "https://en.wikipedia.org/w/index.php?search=%22Data+Provenance+Explorer%22&title=Special:Search&profile=advanced&fulltext=1&ns0=1", "at": "2026-10-29T05:12:06Z", "http": 200, "sha256": "d510d4b8db965748ad8f7fb7fe5406329f363093ef85122c6688c98e0f944658" }, "text": [ "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B11.txt" ], "note": "the API search (list=search) answered HTTP 429 (rate limit) in a probe; the rendered Special:Search page was used instead" }, "names_operator": "no", "names_archive": "no", "note": "no article; the full-text search for the quoted phrase in article space returns no results" }, { "id": "B12", "origin": "wikipedia", "role": "the operator's Wikipedia title", "card": "Wikipedia — Data Provenance Initiative (no article)", "url": "https://en.wikipedia.org/wiki/Data_Provenance_Initiative", "kind": "Wikipedia (no article)", "surfaced_by_T": "no (not surfaced by T; checked by the operator's ruling of 2026-10-29)", "files": { "B12.wiki": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B12.wiki", "sha256": "1d1b5b8029e76a2946ee2990b1cf27dcf3d6f9502c7dc691ea2fd3147410e7c4" }, "B12.html": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B12.html", "sha256": "683496b15564dda466febcf9071448d4c21ebed5af92b66b6aee8ed7fe2af035" }, "B12-search.html": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B12-search.html", "sha256": "8ebe0a384da0592782ff6655bdee204a7210b96179d97c2af8b74bec65399610" }, "B12.txt": { "path": "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B12.txt", "sha256": "979ddc1b561c6122262e6aa8caf4136e2206817fac7fd21ad0023c459ba10827" } }, "fetched": { "raw": { "url": "https://en.wikipedia.org/w/index.php?title=Data_Provenance_Initiative&action=raw&redirect=no", "at": "2026-10-29T05:12:07Z", "http": 404, "sha256": "1d1b5b8029e76a2946ee2990b1cf27dcf3d6f9502c7dc691ea2fd3147410e7c4" }, "rendered": { "url": "https://en.wikipedia.org/wiki/Data_Provenance_Initiative", "at": "2026-10-29T05:12:07Z", "http": 404, "sha256": "683496b15564dda466febcf9071448d4c21ebed5af92b66b6aee8ed7fe2af035" }, "search": { "url": "https://en.wikipedia.org/w/index.php?search=%22Data+Provenance+Initiative%22&title=Special:Search&profile=advanced&fulltext=1&ns0=1", "at": "2026-10-29T05:12:07Z", "http": 200, "sha256": "8ebe0a384da0592782ff6655bdee204a7210b96179d97c2af8b74bec65399610" }, "text": [ "datasets/negative-of-the-negative/v2/intake/data-provenance-explorer-20261029/B12.txt" ], "note": "the API search (list=search) answered HTTP 429 (rate limit) in a probe; the rendered Special:Search page was used instead" }, "names_operator": "no", "names_archive": "no", "note": "no article; the full-text search for the quoted phrase in article space returns no results" } ] } }