{ "row": "ai-hallucination-cases", "entity": "ai-hallucination-cases", "address": "ai hallucination cases database", "addresses": [ { "address": "ai hallucination cases database", "slug": "ai-hallucination-cases-database-aio-20261029", "addr_id": "ADDR-faa512a27b95", "obs_id": "OBS-4f95ab35a3ac", "axn": "AXN:0AA6.CAPTURE.🌋🤙🝊🔙🕐🧬" } ], "epoch": "2026-10-29", "type": "C (public entity on which the archive bears: a database; 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 and the capture record; archive selected by D/R/O (no deposit admitted: D 0, R 0, O-class 10 candidates in 9 deposits read and rejected; no hop), with the MMRS record (26 claims) and the Capture Registry's findings and analysis (6 claims) ledgered as the archive's records of the entity; the field B by three origins (6 cards, 1 own page, Wikipedia: no article and 3 articles naming it); 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 59/59 verified by script against saved text, oEmbed title or card text; archive quotes 32/32 verified against the registry, census.json and articles.py; carried as marked: MMRS-20 contested (F40, F54, F56), MMRS-16, MMRS-24, CAP-1-02 and CAP-1-03 with defects noted", "frozen": false, "breaches": [] }, "spec": "EA-NEGONT-02 v0.9, #1690", "objects": { "T": { "path": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/T1-ai-hallucination-cases-database-aio-20261029.txt", "sha256": "0a0aaa6b60370f0021eef2ac45a942e2a26e0eb37bc4f34d0b77f8e754445901", "words": 174, "cards": 6, "words_note": "one transcript at one address (174 words; 6 cards); kept whole under 'transcripts'", "claims": [ "T1-1 definition: 'The primary public database tracking generative AI hallucinations in court decisions is the AI Hallucination Cases Database, created and maintained by Damien Charlotin.' (markers: card 1's token and one unresolved token)", "T1-2 focus: 'It tracks court and tribunal decisions where generative AI explicitly produced hallucinated content, most commonly fake legal citations, fabricated quotes, and non-existent case law.'", "T1-3 scope and size: 'As of October 2026, the database contains over 2,200 documented decisions globally, with U.S. federal and state courts making up the majority of the cases.'", "T1-4 parties: 'The tracked decisions span private lawyers, pro se (self-represented) litigants, expert witnesses, prosecutors, and judges.'", "T1-5 sanctions: 'The database helps track legal consequences ranging from strict warnings and mandatory verification rules to substantial monetary sanctions and disbarment.'", "T1-6 access: 'You can view and search the project directly at the AI Hallucination Cases Database.' (link and marker: card 1's token)", "T1-7 companion: 'For specialized UK records, a companion resource is available via the AI Hallucination Cases Tracker.' (link and marker: card 2's token)", "T1-8 closer: 'Would you like information on specific high-profile sanctions cases from the database, or guidance on how legal professionals verify AI output?'" ], "register": { "register": "data/EA-WG-CAPTURES-01.json", "slug": "ai-hallucination-cases-database-aio-20261029", "obs_id": "OBS-4f95ab35a3ac" }, "transcripts": [ { "slug": "ai-hallucination-cases-database-aio-20261029", "address": "ai hallucination cases database", "addr_id": "ADDR-faa512a27b95", "obs_id": "OBS-4f95ab35a3ac", "axn": "AXN:0AA6.CAPTURE.🌋🤙🝊🔙🕐🧬", "date": "2026-10-29", "surface": "Google AI Overview (expanded)", "auth": "signed out, incognito", "sha256": "0a0aaa6b60370f0021eef2ac45a942e2a26e0eb37bc4f34d0b77f8e754445901", "sha256_of": "the registry's transcript string (data/EA-WG-CAPTURES-01.json, entries[slug].transcript), UTF-8", "sha256_raw": "83f95e2a97ea881ae465d53a5f2ba08bec5e372b996d5a94d5bf42e0448143f1", "words": 174, "words_note": "words of the composition from the answer's first sentence to its closing question, link targets and citation markers removed; the cards' titles and snippets excluded", "cards": 6, "card_rail": [ "1. Damien Charlotin — AI Hallucination Cases Database", "2. Natural and Artificial Law — AI Hallucination Cases Tracker", "3. HAQQ — AI Hallucination Cases: The 2202-Case Sanctions Tracker - HAQQ", "4. LinkedIn — George Hannah's Post", "5. YouTube·KARE 11 — How common are AI hallucinations in court cases?", "6. YouTube·Miller Law Group — AI Hallucinations Got an Attorney Disbarred" ], "claims": [ "T1-1 definition: 'The primary public database tracking generative AI hallucinations in court decisions is the AI Hallucination Cases Database, created and maintained by Damien Charlotin.' (markers: card 1's token and one unresolved token)", "T1-2 focus: 'It tracks court and tribunal decisions where generative AI explicitly produced hallucinated content, most commonly fake legal citations, fabricated quotes, and non-existent case law.'", "T1-3 scope and size: 'As of October 2026, the database contains over 2,200 documented decisions globally, with U.S. federal and state courts making up the majority of the cases.'", "T1-4 parties: 'The tracked decisions span private lawyers, pro se (self-represented) litigants, expert witnesses, prosecutors, and judges.'", "T1-5 sanctions: 'The database helps track legal consequences ranging from strict warnings and mandatory verification rules to substantial monetary sanctions and disbarment.'", "T1-6 access: 'You can view and search the project directly at the AI Hallucination Cases Database.' (link and marker: card 1's token)", "T1-7 companion: 'For specialized UK records, a companion resource is available via the AI Hallucination Cases Tracker.' (link and marker: card 2's token)", "T1-8 closer: 'Would you like information on specific high-profile sanctions cases from the database, or guidance on how legal professionals verify AI output?'" ], "markers": "inline links are opaque google.com/goto tokens; the definition's first marker, the 'Focus' line's second and the 'Where to Access It' link share card 1's token (the database page); the UK link and its marker share one token, card 2's (Natural and Artificial Law); the definition's second marker and the other markers are unresolved", "path": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/T1-ai-hallucination-cases-database-aio-20261029.txt" } ] }, "KO": { "title": "AI Hallucination Cases", "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": "AI Hallucination Cases is Damien Charlotin's database of legal decisions in which a court or tribunal finds or implies reliance on hallucinated material (MMRS record); its own page says it \"tracks legal decisions\" in cases where generative AI produced hallucinated content, \"typically fake citations, but also other types of AI-generated arguments\", and describes itself as \"the most comprehensive database of AI hallucination cases in law\".", "claims": [ "MMRS-01", "F1", "F22" ] }, { "n": 2, "para": 0, "sense": "maker", "text": "Charlotin wrote in November 2025 that he maintains it on his personal website (FAQ), and the MMRS record gives him as its operator by his own statement; HAQQ's tracker calls him a research fellow at HEC Paris, and Wikipedia's articles on artificial intelligence and on hallucination a legal data analysis lecturer at HEC Paris and Sciences Po.", "claims": [ "F39", "MMRS-13", "F29", "F54", "F56" ] }, { "n": 3, "para": 1, "sense": "the unit", "text": "Its unit is the decision (MMRS record), which its page defines as any document \"where the use of AI, whether established or merely alleged, is addressed in more than a passing reference by the court or tribunal\"; it admits \"only cases where the court or tribunal has explicitly found (or implied) that a party relied on hallucinated content or material\", which the MMRS record states as reliance a court \"explicitly finds, or implies\", entering it in its self-description field and as the property that admission requires \"a court finding or implication of reliance on hallucinated material\".", "claims": [ "MMRS-02", "F2", "F3", "MMRS-03", "MMRS-16", "MMRS-21" ] }, { "n": 4, "para": 1, "sense": "admission", "text": "The judgment is the courts': \"I am not making that judgment, I let the courts and judges make or imply it, which is why the database is necessarily an undercount\" (FAQ); HAQQ restates the bar as a court that \"found, or clearly implied\" reliance, accusations alone not counting, and the FAQ holds that most hallucinations are \"very obvious\", a made-up case name or false quote one that \"cannot be a human mistake\".", "claims": [ "F44", "F29", "F46" ] }, { "n": 5, "para": 1, "sense": "exceptions", "text": "Its exceptions are stated twice, and the statements pull apart: its page \"also covers some judicial decisions where AI use was alleged but not confirmed\", \"a judgment call on my part\", while the FAQ says he refrains \"from adding rows about cases where hallucinations are only alleged\" and, in a February 2026 edit, makes an exception where the alleged hallucination comes from a judge and, absent an appeal decision or retraction, he is \"bound to make a judgment call\"; the MMRS record composes both as \"Mere allegations are excluded, with some judgment calls admitted.\"", "claims": [ "F4", "F45", "F47", "MMRS-04" ] }, { "n": 6, "para": 2, "sense": "what its selection leaves out", "text": "It states its boundary itself, that it \"does not track the (necessarily wider) universe of all fake citations or use of AI in court filings\" (its page; quoted by the MMRS record), and on the MMRS record's reading a fabricated citation enters only when a court catches and rules on it; HAQQ draws the same consequence for its reading of the count, \"a fake citation nobody checks never reaches the database, so 2,202 is a floor\".", "claims": [ "F5", "MMRS-09", "MMRS-10", "F30" ] }, { "n": 7, "para": 3, "sense": "the record", "text": "Each record gives the case, the court and jurisdiction, the date, the party using AI, the AI tool, the nature of the hallucination, the outcome or sanction, the monetary penalty, details and reports (its table; the MMRS record lists the court, date, party, tool where named, nature and outcome), the nature filed as Fabricated (1,829), False Quotes (595), Misrepresented (947) or Outdated Advice (38).", "claims": [ "F17", "MMRS-05", "F15" ] }, { "n": 8, "para": 3, "sense": "the record", "text": "Its outcomes can be filtered for monetary sanctions and disciplinary referrals, and a penalty recorded as 1 \"is a placeholder, not the real figure\", the amount being unknown (its page).", "claims": [ "F23", "F21" ] }, { "n": 9, "para": 3, "sense": "the record's limits", "text": "Its page states the record's limits: AI tools are tracked \"only as disclosed or admitted by parties in court filings\", with 'Unidentified' and 'Implied' as the defaults; a named tool \"does not necessarily mean that tool was responsible\"; offending-document dates are parsed automatically and \"may occasionally be wrong\", and responses when caught are extracted automatically, \"indicative rather than exact\".", "claims": [ "F18", "F19", "F20" ] }, { "n": 10, "para": 4, "sense": "count", "text": "Seeking to be exhaustive, it had identified 2,231 cases at its update of 28 October 2026 and calls itself \"a work in progress\" (its page; the MMRS record gives the same count as of the site's own update date), 1,514 of them from the USA, 238 from Canada, 112 from Australia and 82 from the UK, with 1,288 involving pro se litigants, 880 lawyers, 32 judges, 17 experts and 5 prosecutors (its filters; the MMRS record gives the pro se and lawyer counts).", "claims": [ "F6", "MMRS-18", "F14", "F13", "MMRS-06" ] }, { "n": 11, "para": 4, "sense": "count", "text": "HAQQ, reading the database's update of 19 October 2026, gives 2,202 decisions, 1,501 from the United States, 1,267 involving pro se litigants and 872 lawyers.", "claims": [ "F28", "F31" ] }, { "n": 12, "para": 5, "sense": "origin", "text": "Charlotin started it in April 2025 while teaching Sciences Po Law School students on the \"Limits and Potentials\" of LLMs in law, when he looked for data on hallucinations and, finding none, decided to \"tally the cases myself\" (FAQ); Wikipedia dates its establishment to April 2025 and places its creation among the effects of law firms' use of generative AI for research; its period filter begins in the second quarter of 2023 (its page; MMRS record), and the MMRS record's census enters that filter as its founding.", "claims": [ "F40", "F54", "F56", "F16", "MMRS-07", "MMRS-20" ] }, { "n": 13, "para": 5, "sense": "growth", "text": "Before 2025, the FAQ says, \"we maybe had two or three cases a month\", and by February 2026 the daily average was five; Wikipedia's hallucination article gives over 1,300 instances a year after its start, a LinkedIn post four months before the capture 1,600 cases, and HAQQ's series about 200 in mid-2025, 719 in January 2026 and 2,202 on 19 October 2026; its page reads \"Updated daily\" and was last updated on 28 October 2026, and the MMRS record gives its cadence as continuous.", "claims": [ "F41", "F56", "F36", "F32", "F22", "F11", "MMRS-19" ] }, { "n": 14, "para": 5, "sense": "causes and coverage", "text": "The FAQ attributes the acceleration to judicial lag, the availability of LLMs and public knowledge of them, matching a profile of self-represented litigants or lawyers surprised by a tool's new AI component, and explains the share of US entries by the US legal data ecosystem, legal data in many European countries being \"hoarded by legal editors\".", "claims": [ "F42", "F48" ] }, { "n": 15, "para": 5, "sense": "causes and coverage", "text": "It divides the records into pure mistakes, \"the vast majority of cases\", and a \"(substantial) minority\" involving \"bad\" actors (FAQ).", "claims": [ "F49" ] }, { "n": 16, "para": 6, "sense": "intake", "text": "Cases come from referrals, scrapers and bots monitoring data sources, and keyword searches (FAQ), and readers are asked to send cases (its page); the MMRS record gives the intake as editor-curated with readers sending cases and the database growing by its editor's reading and readers' cases; the pipeline for new entries itself uses AI, \"though I am careful to check it does not hallucinate\" (FAQ).", "claims": [ "F43", "F9", "MMRS-17", "MMRS-08", "F50" ] }, { "n": 17, "para": 6, "sense": "openness", "text": "It is licensed CC BY 4.0, \"free to reuse, including commercially, with attribution\", to be cited as \"AI Hallucination Cases Database, Damien Charlotin\", and its cases download as CSV (its page; MMRS record).", "claims": [ "F12", "F11", "MMRS-15" ] }, { "n": 18, "para": 7, "sense": "purpose", "text": "The FAQ gives its use as practitioners finding the cases relevant in their own jurisdiction, data analyses, and the cases as \"benchmarks to fix the issue\"; from it Charlotin built PelAIkan, \"an automated reference checker that also detects hallucinations\" (its page), and he expects no fix \"at the model stage\" and no reduction to zero \"given the existing paradigm\" (FAQ, a forecast).", "claims": [ "F51", "F10", "F52" ] }, { "n": 19, "para": 8, "sense": "citation fidelity", "text": "\"For years, we (me included) have cited without reading; but now the costs of that practice have become explicit\": hallucinations, the FAQ says, \"expose the epistemic hygiene the legal profession has long lacked\".", "claims": [ "F53" ] }, { "n": 20, "para": 8, "sense": "citation fidelity", "text": "The MMRS record reads it as \"the census's most exact record of citation fidelity\", recording attribution by the nature of the hallucination: \"Each entry is a machine composition that reached a court with sources it invented or misrepresented, and the court's decision is the reception that recorded it.\"", "claims": [ "MMRS-11", "MMRS-22", "MMRS-12" ] }, { "n": 21, "para": 9, "sense": "reception", "text": "It has been featured in news media \"and indeed in several decisions dealing with hallucinated material\", the LA Times among them in May 2025 (its page); Wikipedia cites it for counts of Canadian filings (at least 167 by 14 June 2026, with courtready.ca) and Israeli cases (at least 55 by August 2026) and, in its article on Mata v. Avianca, for that case's description as a landmark, and its article on artificial intelligence records judges' sanctions and bar warnings by 2026.", "claims": [ "F7", "F8", "F57", "F58", "F59", "F55" ] }, { "n": 22, "para": 9, "sense": "reception", "text": "On the Overview's video cards it appears at title grain: KARE 11's \"How common are AI hallucinations in court cases?\" and Miller Law Group's \"AI Hallucinations Got an Attorney Disbarred\".", "claims": [ "F37", "F38" ] }, { "n": 23, "para": 10, "sense": "neighbours", "text": "HAQQ, which discloses that it \"sells a legal AI product\", re-reads the live database monthly and takes off its page any number it cannot re-verify, and lists sanctions from revoked admissions to personal costs, the largest it can source about $109,700 (HAQQ); the MMRS record enters it as a candidate, \"a vendor tracker built on Charlotin's count\".", "claims": [ "F35", "F34", "F33", "MMRS-25" ] }, { "n": 24, "para": 10, "sense": "neighbours", "text": "Matthew Lee, a barrister at Doughty Street Chambers, keeps at Natural and Artificial Law an international tracker of \"Selected judicial quotations\" on AI hallucination cases and fabricated or false citations, first published 16 May 2025, which \"does not measure how often these issues occur\", and a separate UK tracker of case records, outcomes, quotations and sources in which \"inclusion is not proof of an AI hallucination or wrongdoing\"; neither page names Charlotin's database, which the MMRS record's candidate list enters as \"a UK companion to Charlotin's database\".", "claims": [ "F26", "F27", "F24", "F25", "MMRS-24" ] }, { "n": 25, "para": 10, "sense": "neighbours", "text": "The Capture Registry (2026-10-29) holds the UK tracker a separate aggregator and HAQQ's tracker, which gives the database's 2,202 decisions of 19 October (HAQQ), built on Charlotin's count, both added as candidates the census did not hold, and reads the cards as the database \"already aggregated in turn: a regional companion and a vendor tracker counting its cases\".", "claims": [ "CAP-1-03", "F28", "CAP-1-06" ] }, { "n": 26, "para": 11, "sense": "the record kept apart", "text": "The MMRS record keeps itself apart from the entity, each assertion with a source and a basis and the lifecycle apart from the observation that saw it, records the database as active on 28 October 2026 with the roles of monitoring and reception, and relates it to the composition layer's account seated in the Capture Registry.", "claims": [ "MMRS-26", "MMRS-14", "MMRS-23" ] }, { "n": 27, "para": 11, "sense": "the composition layer", "text": "There (Capture Registry, 2026-10-29) the Overview at 'ai hallucination cases database' composes the database with its maker and at its size, Charlotin and 'over 2,200' being right against the 2,231 its page gives, as the field's primary record; it narrows the admission rule, which the page states as reliance the court \"has explicitly found (or implied)\", to 'explicitly', so that the implied cases are dropped, and lists private lawyers first where the page's filters give 1,288 pro se litigants and 880 lawyers.", "claims": [ "CAP-1-01", "F6", "CAP-1-05", "CAP-1-02", "F3", "CAP-1-04", "F13" ] } ], "joins": [ { "act": "order of senses", "sentences": "1–27", "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 and its maker (¶0) precede the unit and the admission rule (¶1), on which the boundary (¶2), the record and its limits (¶3) and the count (¶4) depend; origin, growth and the maker's reading of the cases (¶5) follow the count they date; intake and openness (¶6) and purpose (¶7) follow; citation fidelity (¶8) reads the record already composed; reception (¶9), the neighbours (¶10) and the record kept apart with the composition layer's observation (¶11) stand last, as relations of the entity already composed." }, { "act": "holding apart (two statements of the exceptions)", "sentences": [ 5 ], "claims": [ "F4", "F45", "F47", "MMRS-04" ], "note": "P_coh: the page's admission of some alleged-but-unconfirmed decisions and the FAQ's refusal of rows only alleged, with its judge exception, composed each with its source; the MMRS record's composite stated as its own; no coordinating claim of compatibility (open block)" }, { "act": "holding apart (contested founding)", "sentences": [ 12 ], "claims": [ "F40", "F54", "F56", "F16", "MMRS-07", "MMRS-20" ], "note": "P_coh: April 2025 (FAQ; Wikipedia) and the MMRS census's 'founded' (the period filter, Q2 2023) composed each with its source; the period filter itself is a field fact (F16)" }, { "act": "holding together (count at two dates)", "sentences": [ 10, 11 ], "claims": [ "F6", "MMRS-18", "F28", "F31" ], "note": "the page's 2,231 (28 October) and HAQQ's 2,202 (19 October) composed as readings at different dates, each dated" }, { "act": "holding together (the boundary and its consequence)", "sentences": [ 6 ], "claims": [ "F5", "MMRS-09", "MMRS-10", "F30" ], "note": "the page's own boundary, the MMRS record's reading of what enters, and HAQQ's 'floor' composed as one sense: what the selection leaves out" }, { "act": "holding apart (companion or neighbour)", "sentences": [ 24, 25 ], "claims": [ "F24", "F26", "MMRS-24", "CAP-1-03" ], "note": "the MMRS candidate's 'UK companion to Charlotin's database' and the Capture Registry's 'separate aggregator' beside Natural and Artificial Law's pages, which do not name the database; composed each with its source (open block)" } ], "plan": { "order": [ "MMRS-01", "F1", "F22", "F39", "MMRS-13", "F29", "F54", "F56", "MMRS-02", "F2", "F3", "MMRS-03", "MMRS-16", "MMRS-21", "F44", "F29", "F46", "F4", "F45", "F47", "MMRS-04", "F5", "MMRS-09", "MMRS-10", "F30", "F17", "MMRS-05", "F15", "F23", "F21", "F18", "F19", "F20", "F6", "MMRS-18", "F14", "F13", "MMRS-06", "F28", "F31", "F40", "F54", "F56", "F16", "MMRS-07", "MMRS-20", "F41", "F56", "F36", "F32", "F22", "F11", "MMRS-19", "F42", "F48", "F49", "F43", "F9", "MMRS-17", "MMRS-08", "F50", "F12", "F11", "MMRS-15", "F51", "F10", "F52", "F53", "MMRS-11", "MMRS-22", "MMRS-12", "F7", "F8", "F57", "F58", "F59", "F55", "F37", "F38", "F35", "F34", "F33", "MMRS-25", "F26", "F27", "F24", "F25", "MMRS-24", "CAP-1-03", "F28", "CAP-1-06", "MMRS-26", "MMRS-14", "MMRS-23", "CAP-1-01", "F6", "CAP-1-05", "CAP-1-02", "F3", "CAP-1-04", "F13" ], "senses": { "what it is": [ 1 ], "maker": [ 2 ], "the unit": [ 3 ], "admission": [ 4 ], "exceptions": [ 5 ], "what its selection leaves out": [ 6 ], "the record": [ 7, 8 ], "the record's limits": [ 9 ], "count": [ 10, 11 ], "origin": [ 12 ], "growth": [ 13 ], "causes and coverage": [ 14, 15 ], "intake": [ 16 ], "openness": [ 17 ], "purpose": [ 18 ], "citation fidelity": [ 19, 20 ], "reception": [ 21, 22 ], "neighbours": [ 23, 24, 25 ], "the record kept apart": [ 26 ], "the composition layer": [ 27 ] }, "case_by_sense": { "equal (one claim space; nothing translated)": [ "what it is", "maker", "the unit", "admission", "exceptions", "what its selection leaves out", "the record", "the record's limits", "count", "origin", "growth", "causes and coverage", "intake", "openness", "purpose", "citation fidelity", "reception", "neighbours", "the record kept apart", "the composition layer" ] }, "carried": { "archive": 32, "field": 59, "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 347 words) with no channel log" }, { "rule": "P_coh", "result": "F4 against F45/F47 on alleged cases (sentence 5); MMRS-20 against F40/F54/F56 on the founding (sentence 12); MMRS-24 against F24/F26 on the companion relation (sentence 24): each composed with its source" }, { "rule": "N_c", "result": "no claim lowered by standing; the MMRS record's readings (MMRS-10, MMRS-11, MMRS-12, MMRS-22, MMRS-24, MMRS-25) carried at interpretation, the FAQ's forecast (F52) at hypothesis" }, { "rule": "C_ex", "result": "senses composed ⊇ senses in the ledgers: every ledger lineage cited in the KO" } ], "marks": { "contested": [ "MMRS-20 (founded: the period filter against April 2025)" ], "defects": [ "MMRS-16: the rendered MMRS page shows this field under 'In its own words', in quotation marks; the words are the record's compression of the page's footnote 1 and opening (F1–F3), not a sentence of the database's page", "MMRS-24: the card's page (naturalandartificiallaw.com/ai-hallucination-cases-tracker/) is Matthew Lee's international tracker of selected judicial quotations, which links to a separate UK tracker (/uk-ai-hallucination-cases-tracker/); neither page in the saved text names Charlotin or his database; 'companion' is the Overview's word (T1-7); the candidate's url is null", "CAP-1-02: the phrase quoted as the database's rule, 'explicitly finds, or implies', is the MMRS article's wording (MMRS-03); the page reads 'has explicitly found (or implied)' (F3); the substance agrees", "CAP-1-03: the card (B2) is the international page of Natural and Artificial Law's trackers; the UK tracker is a further page it links to, not on the rail" ] }, "cuts": [] } }, "P": { "title": "AI Hallucination Cases", "lede": { "text": "is Damien Charlotin's database of legal decisions in which a court or tribunal finds or implies reliance on hallucinated material (MMRS record; its page): 2,231 cases as of 28 October 2026, CC BY 4.0, with a CSV export.", "claims": [ "MMRS-01", "MMRS-13", "F3", "F6", "F12", "F11", "MMRS-15" ] }, "sections": [ { "icon": "📋", "head": "What it records", "items": [ { "label": "Unit:", "text": "the decision, any document addressing AI use beyond a passing reference; typically fake citations (its page; MMRS record).", "claims": [ "MMRS-02", "F2", "F1", "MMRS-16" ] }, { "label": "Admission:", "text": "the courts' finding or implication, so 'necessarily an undercount' (FAQ); some alleged decisions admitted as judgment calls, against the FAQ's refusal of rows only alleged.", "claims": [ "F3", "F44", "MMRS-03", "MMRS-21", "F4", "F45", "F47", "MMRS-04" ] }, { "label": "Fields:", "text": "court, party, tool, nature (fabricated, false quotes, misrepresented, outdated advice), outcome, penalty; tools only as disclosed, not necessarily responsible.", "claims": [ "F17", "MMRS-05", "F15", "F23", "F18", "F19" ] } ] }, { "icon": "📈", "head": "Size and growth", "items": [ { "label": "Counts:", "text": "USA 1,514; pro se 1,288, lawyers 880 (its filters; MMRS record); HAQQ read 2,202 on 19 October.", "claims": [ "F14", "F13", "MMRS-06", "MMRS-18", "F28", "F31" ] }, { "label": "Growth:", "text": "two or three a month before 2025, five a day by February 2026 (FAQ); continuous (MMRS record).", "claims": [ "F41", "MMRS-19" ] }, { "label": "Origin:", "text": "April 2025, a Sciences Po course (FAQ; Wikipedia); the MMRS census enters the period filter's Q2 2023 as its founding.", "claims": [ "F40", "F54", "F16", "MMRS-07", "MMRS-20" ] } ] }, { "icon": "🧭", "head": "Boundary and use", "items": [ { "label": "Not tracked:", "text": "'the (necessarily wider) universe of all fake citations or use of AI in court filings'; a citation enters when a court catches it (MMRS record); a floor (HAQQ).", "claims": [ "F5", "MMRS-09", "MMRS-10", "F30" ] }, { "label": "Intake:", "text": "referrals, scrapers, searches, readers' cases, a checked AI pipeline (FAQ; MMRS record).", "claims": [ "F43", "F9", "F50", "MMRS-08", "MMRS-17" ] }, { "label": "Uses:", "text": "jurisdiction research, benchmarks; PelAIkan, a checker built on it; no fix expected at the model stage (FAQ).", "claims": [ "F51", "F10", "F52" ] } ] }, { "icon": "🔗", "head": "Citation fidelity", "items": [ { "label": "Archive:", "text": "the census's most exact record of citation fidelity: each entry 'a machine composition that reached a court with sources it invented or misrepresented' (MMRS record).", "claims": [ "MMRS-11", "MMRS-12", "MMRS-22" ] }, { "label": "Maker:", "text": "'we (me included) have cited without reading' (FAQ).", "claims": [ "F53" ] } ] }, { "icon": "🗂️", "head": "Around it", "items": [ { "label": "HAQQ:", "text": "a legal-AI vendor's monthly re-count; sanctions to about $109,700 (HAQQ; MMRS candidate).", "claims": [ "F34", "F35", "F33", "MMRS-25" ] }, { "label": "Natural and Artificial Law:", "text": "Matthew Lee's judicial quotations and UK tracker; neither names the database, which the MMRS candidates call its UK companion.", "claims": [ "F26", "F27", "F24", "F25", "MMRS-24" ] }, { "label": "Reception:", "text": "news media, court decisions, Wikipedia, Mata v. Avianca; two videos by title.", "claims": [ "F7", "F8", "F55", "F57", "F58", "F59", "F37", "F38" ] } ] }, { "icon": "👁️", "head": "On the Overview", "items": [ { "label": "Recorded:", "text": "maker and size right; 'finds or implies' narrowed to 'explicitly'; lawyers before the larger pro se count (Capture Registry, 2026-10-29).", "claims": [ "CAP-1-01", "CAP-1-02", "CAP-1-05", "CAP-1-04" ] }, { "label": "Record:", "text": "kept apart from the entity, each field with source and basis; active (MMRS record).", "claims": [ "MMRS-26", "MMRS-14", "MMRS-23" ] } ] } ], "open": [ { "question": "Does it admit decisions where AI use or hallucination was alleged but not confirmed?", "rests_on": [ "F3", "F4", "F45", "F47", "MMRS-04" ], "modality": "stipulation (its page and its FAQ, two statements of the maker that pull apart); attributed (MMRS record)", "would_resolve": "a flag in the CSV marking alleged-only entries, or the maker's statement of which rule governs" }, { "question": "When did it begin: April 2025, or the second quarter of 2023?", "rests_on": [ "F40", "F54", "F56", "F16", "MMRS-07", "MMRS-20" ], "modality": "documented (FAQ; Wikipedia) against documented (the MMRS census's 'founded', basis observed, which holds the period filter)", "would_resolve": "the MMRS record's own correction: the period filter dates decisions, the FAQ dates the database" }, { "question": "How much wider is the universe it does not track?", "rests_on": [ "F5", "F44", "F30", "MMRS-10" ], "modality": "self-description; stipulation; interpretation", "would_resolve": "a count of fabricated citations in a sample of filings, decided or not, against the database's entries for the same courts" }, { "question": "Is Natural and Artificial Law's tracker a companion to the database?", "rests_on": [ "F24", "F26", "MMRS-24", "CAP-1-03" ], "modality": "documented (Natural and Artificial Law, which does not name the database) against interpretation (MMRS candidate)", "would_resolve": "a statement by Matthew Lee or Charlotin of a relation between the trackers" }, { "question": "Opacities: the maker's title (research fellow at HEC Paris, or HEC Paris and Sciences Po lecturer); the LinkedIn card unreachable, its snippet the only text; two videos by title only.", "rests_on": [ "F29", "F54", "F56", "F40", "F36", "F37", "F38" ], "modality": "attributed; documented", "would_resolve": "the maker's own page or CV; the post and the videos' transcripts" } ], "body_words": 347, "rail": [ { "lineage": "what it is: legal decisions on AI hallucinations", "claims": [ "F1", "F22", "MMRS-01", "MMRS-16" ], "earliest": "B1" }, { "lineage": "maker and origin", "claims": [ "F16", "F39", "F40", "F54", "F56", "MMRS-07", "MMRS-13", "MMRS-20" ], "earliest": "B7" }, { "lineage": "the unit and the admission rule", "claims": [ "F2", "F3", "F29", "F44", "F46", "MMRS-02", "MMRS-03", "MMRS-21" ], "earliest": "B1" }, { "lineage": "exceptions: judgment calls", "claims": [ "F4", "F45", "F47", "MMRS-04" ], "earliest": "B1" }, { "lineage": "boundary: what it does not track", "claims": [ "F5", "F30", "MMRS-09", "MMRS-10" ], "earliest": "B1" }, { "lineage": "the record: fields and categories", "claims": [ "F15", "F17", "F21", "F23", "MMRS-05" ], "earliest": "B1" }, { "lineage": "the record's stated limits", "claims": [ "F18", "F19", "F20" ], "earliest": "B1" }, { "lineage": "count and composition", "claims": [ "F6", "F13", "F14", "F28", "F31", "MMRS-06", "MMRS-18" ], "earliest": "B1" }, { "lineage": "growth", "claims": [ "F32", "F41", "MMRS-19" ], "earliest": "B7" }, { "lineage": "causes and coverage: the maker's reading", "claims": [ "F42", "F48", "F49" ], "earliest": "B7" }, { "lineage": "intake and pipeline", "claims": [ "F9", "F43", "F50", "MMRS-08", "MMRS-17" ], "earliest": "B7" }, { "lineage": "openness: licence and data", "claims": [ "F11", "F12", "MMRS-15" ], "earliest": "B1" }, { "lineage": "purpose and uses", "claims": [ "F10", "F51", "F52" ], "earliest": "B7" }, { "lineage": "citation fidelity", "claims": [ "F53", "MMRS-11", "MMRS-12", "MMRS-22" ], "earliest": "B7" }, { "lineage": "reception: press, courts, encyclopedia, video", "claims": [ "F7", "F8", "F36", "F37", "F38", "F55", "F57", "F58", "F59" ], "earliest": "B1" }, { "lineage": "neighbour: HAQQ's sanctions tracker", "claims": [ "F33", "F34", "F35", "MMRS-25" ], "earliest": "B3" }, { "lineage": "neighbour: Natural and Artificial Law's trackers", "claims": [ "F24", "F25", "F26", "F27", "MMRS-24" ], "earliest": "B2" }, { "lineage": "the record kept apart from the entity", "claims": [ "MMRS-14", "MMRS-23", "MMRS-26" ], "earliest": "MMRS" }, { "lineage": "reception: the composition layer", "claims": [ "CAP-1-01", "CAP-1-02", "CAP-1-03", "CAP-1-04", "CAP-1-05", "CAP-1-06" ], "earliest": "CAP" } ], "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, with AIO's transcript a record within it", "status": "composed 2026-10-29 from the v0.9 knowledge object; not frozen", "entity": "Composed under v0.9, equal case (ρ = extends): KO_B sentences surviving 0 of 10; every field claim carried." }, "P_B": { "title": "AI Hallucination Cases", "arm": "B (the field alone)", "genre": "popup — the compression from the field ledger alone (cards, own page, Wikipedia)", "status": "composed 2026-10-29 from the field ledger alone; not frozen", "lede": { "text": "is a database that tracks legal decisions in which a court or tribunal has explicitly found (or implied) that a party relied on AI-hallucinated content, typically fake citations (its page): 2,231 cases as of 28 October 2026.", "claims": [ "F1", "F3", "F6" ] }, "sections": [ { "icon": "📋", "head": "What it records", "items": [ { "label": "Admission:", "text": "the courts decide, so it undercounts (FAQ); some alleged decisions admitted as judgment calls (its page).", "claims": [ "F44", "F4", "F47" ] }, { "label": "Fields:", "text": "court, party, tool as disclosed, nature, outcome, penalty (its page).", "claims": [ "F17", "F15", "F18", "F23" ] } ] }, { "icon": "📈", "head": "Size and origin", "items": [ { "label": "Counts:", "text": "USA 1,514; pro se 1,288, lawyers 880; HAQQ 2,202 (19 October).", "claims": [ "F14", "F13", "F28", "F31" ] }, { "label": "Origin:", "text": "Charlotin, April 2025, a Sciences Po course (FAQ; Wikipedia).", "claims": [ "F40", "F39", "F54", "F56", "F16" ] }, { "label": "Growth:", "text": "five a day by February 2026 (FAQ).", "claims": [ "F41" ] }, { "label": "Boundary:", "text": "not the wider universe of fake citations; a floor (its page; HAQQ).", "claims": [ "F5", "F30" ] } ] }, { "icon": "🗂️", "head": "Around it", "items": [ { "label": "Use:", "text": "CC BY 4.0, CSV; practitioners, benchmarks, PelAIkan (its page; FAQ).", "claims": [ "F12", "F11", "F51", "F10" ] }, { "label": "Neighbours:", "text": "HAQQ's re-count; Matthew Lee's judicial quotations and UK tracker.", "claims": [ "F34", "F26", "F24" ] }, { "label": "Reception:", "text": "news, decisions, Wikipedia, two videos.", "claims": [ "F7", "F8", "F57", "F58", "F59", "F37", "F38" ] } ] } ], "open": [ { "question": "Does it admit decisions where hallucination was only alleged?", "rests_on": [ "F3", "F4", "F45", "F47" ], "modality": "stipulation (two statements of the maker)", "would_resolve": "the maker's statement of which rule governs" }, { "question": "Opacities: the maker's title; the LinkedIn card unreachable; two videos by title only.", "rests_on": [ "F29", "F54", "F36", "F37", "F38" ], "modality": "attributed; documented", "would_resolve": "the maker's CV; the post and video transcripts" } ], "entity": "With no archive claim admitted, the entry is the field's database at its own resolution: its page and FAQ, its neighbours' trackers, the encyclopedia's sections, and the cards.", "body_words": 140, "rail": [ { "lineage": "what it is: legal decisions on AI hallucinations", "claims": [ "F1", "F22" ], "earliest": "B1" }, { "lineage": "maker and origin", "claims": [ "F16", "F39", "F40", "F54", "F56" ], "earliest": "B7" }, { "lineage": "the unit and the admission rule", "claims": [ "F2", "F3", "F29", "F44", "F46" ], "earliest": "B1" }, { "lineage": "exceptions: judgment calls", "claims": [ "F4", "F45", "F47" ], "earliest": "B1" }, { "lineage": "boundary: what it does not track", "claims": [ "F5", "F30" ], "earliest": "B1" }, { "lineage": "the record: fields and categories", "claims": [ "F15", "F17", "F21", "F23" ], "earliest": "B1" }, { "lineage": "the record's stated limits", "claims": [ "F18", "F19", "F20" ], "earliest": "B1" }, { "lineage": "count and composition", "claims": [ "F6", "F13", "F14", "F28", "F31" ], "earliest": "B1" }, { "lineage": "growth", "claims": [ "F32", "F41" ], "earliest": "B7" }, { "lineage": "causes and coverage: the maker's reading", "claims": [ "F42", "F48", "F49" ], "earliest": "B7" }, { "lineage": "intake and pipeline", "claims": [ "F9", "F43", "F50" ], "earliest": "B7" }, { "lineage": "openness: licence and data", "claims": [ "F11", "F12" ], "earliest": "B1" }, { "lineage": "purpose and uses", "claims": [ "F10", "F51", "F52" ], "earliest": "B7" }, { "lineage": "citation fidelity", "claims": [ "F53" ], "earliest": "B7" }, { "lineage": "reception: press, courts, encyclopedia, video", "claims": [ "F7", "F8", "F36", "F37", "F38", "F55", "F57", "F58", "F59" ], "earliest": "B1" }, { "lineage": "neighbour: HAQQ's sanctions tracker", "claims": [ "F33", "F34", "F35" ], "earliest": "B3" }, { "lineage": "neighbour: Natural and Artificial Law's trackers", "claims": [ "F24", "F25", "F26", "F27" ], "earliest": "B2" } ] }, "KO_B": { "title": "AI Hallucination Cases", "arm": "B (the field alone)", "genre": "knowledge object", "status": "composed 2026-10-29; not frozen", "sentences": [ { "n": 1, "para": 0, "text": "AI Hallucination Cases tracks legal decisions in cases where generative AI produced hallucinated content, typically fake citations, describing itself as the most comprehensive database of AI hallucination cases in law, and its decision is any document in which a court or tribunal addresses AI use in more than a passing reference (its page).", "claims": [ "F1", "F22", "F2" ] }, { "n": 2, "para": 0, "text": "It covers only cases where the court or tribunal has explicitly found (or implied) reliance on hallucinated material, the maker letting the courts make or imply the judgment so that the database is necessarily an undercount (its page; FAQ; HAQQ), with some alleged but unconfirmed decisions, and alleged hallucinations of judges, as judgment calls, though the FAQ says he refrains from rows only alleged, most hallucinations being obvious.", "claims": [ "F3", "F44", "F29", "F4", "F45", "F47", "F46" ] }, { "n": 3, "para": 1, "text": "Damien Charlotin started it in April 2025 while teaching at Sciences Po and maintains it on his personal website (FAQ), and Wikipedia gives its establishment in April 2025 by the HEC Paris and Sciences Po lecturer; its period filter begins in Q2 2023.", "claims": [ "F40", "F39", "F54", "F56", "F16" ] }, { "n": 4, "para": 1, "text": "It does not track the wider universe of all fake citations or AI use in court filings (its page), and HAQQ calls its count a floor.", "claims": [ "F5", "F30" ] }, { "n": 5, "para": 2, "text": "It had identified 2,231 cases by 28 October 2026, 1,514 from the USA and 1,288 involving pro se litigants against 880 lawyers, with HAQQ's 2,202 on 19 October, 1,501 American; the nature of each hallucination is fabricated, false quotes, misrepresented or outdated advice (its filters; HAQQ).", "claims": [ "F6", "F14", "F13", "F28", "F31", "F15" ] }, { "n": 6, "para": 2, "text": "Each record carries the court, date, party, AI tool as disclosed, nature, outcome, penalty and reports, a penalty of 1 being a placeholder, and the page warns that a named tool is not necessarily responsible and that automatically extracted dates and responses are indicative (its page).", "claims": [ "F17", "F23", "F21", "F18", "F19", "F20" ] }, { "n": 7, "para": 3, "text": "From two or three cases a month before 2025 it reached five a day by February 2026 (FAQ), with over 1,300 a year after its start (Wikipedia) and HAQQ's series from about 200 to 2,202, updated daily (its page); the FAQ reads the growth through judicial lag and LLM availability, the US share through its legal data, and the records as mostly pure mistakes.", "claims": [ "F41", "F56", "F32", "F11", "F42", "F48", "F49", "F36" ] }, { "n": 8, "para": 3, "text": "Cases come from referrals, scrapers and keyword searches, and from readers, through an AI-assisted pipeline (FAQ; its page); it is CC BY 4.0 with a CSV, serves practitioners and benchmarks, underlies the PelAIkan checker, and its maker expects no fix at the model stage and finds that hallucinations expose a profession that cited without reading (FAQ).", "claims": [ "F43", "F9", "F50", "F12", "F51", "F10", "F52", "F53" ] }, { "n": 9, "para": 4, "text": "It has been featured in news media and in court decisions (its page), and Wikipedia cites it for Canadian and Israeli counts and Mata v. Avianca, records sanctions by 2026, and the Overview's cards include a LinkedIn post at 1,600 cases and two videos by title.", "claims": [ "F7", "F8", "F57", "F58", "F59", "F55", "F37", "F38" ] }, { "n": 10, "para": 4, "text": "Beside it stand HAQQ's sanctions tracker, a vendor's monthly re-count that lists sanctions up to about $109,700, and Matthew Lee's international tracker of selected judicial quotations at Natural and Artificial Law, which measures no frequency, with a separate UK tracker whose inclusion is not proof (HAQQ; Natural and Artificial Law).", "claims": [ "F33", "F34", "F35", "F26", "F27", "F24", "F25" ] } ] }, "L_B": { "words": 508, "text": "AI Hallucination Cases tracks legal decisions in cases where generative AI produced hallucinated content, typically fake citations, describing itself as the most comprehensive database of AI hallucination cases in law, and its decision is any document in which a court or tribunal addresses AI use in more than a passing reference (its page). It covers only cases where the court or tribunal has explicitly found (or implied) reliance on hallucinated material, the maker letting the courts make or imply the judgment so that the database is necessarily an undercount (its page; FAQ; HAQQ), with some alleged but unconfirmed decisions, and alleged hallucinations of judges, as judgment calls, though the FAQ says he refrains from rows only alleged, most hallucinations being obvious. Damien Charlotin started it in April 2025 while teaching at Sciences Po and maintains it on his personal website (FAQ), and Wikipedia gives its establishment in April 2025 by the HEC Paris and Sciences Po lecturer; its period filter begins in Q2 2023. It does not track the wider universe of all fake citations or AI use in court filings (its page), and HAQQ calls its count a floor. It had identified 2,231 cases by 28 October 2026, 1,514 from the USA and 1,288 involving pro se litigants against 880 lawyers, with HAQQ's 2,202 on 19 October, 1,501 American; the nature of each hallucination is fabricated, false quotes, misrepresented or outdated advice (its filters; HAQQ). Each record carries the court, date, party, AI tool as disclosed, nature, outcome, penalty and reports, a penalty of 1 being a placeholder, and the page warns that a named tool is not necessarily responsible and that automatically extracted dates and responses are indicative (its page). From two or three cases a month before 2025 it reached five a day by February 2026 (FAQ), with over 1,300 a year after its start (Wikipedia) and HAQQ's series from about 200 to 2,202, updated daily (its page); the FAQ reads the growth through judicial lag and LLM availability, the US share through its legal data, and the records as mostly pure mistakes. Cases come from referrals, scrapers and keyword searches, and from readers, through an AI-assisted pipeline (FAQ; its page); it is CC BY 4.0 with a CSV, serves practitioners and benchmarks, underlies the PelAIkan checker, and its maker expects no fix at the model stage and finds that hallucinations expose a profession that cited without reading (FAQ). It has been featured in news media and in court decisions (its page), and Wikipedia cites it for Canadian and Israeli counts and Mata v. Avianca, records sanctions by 2026, and the Overview's cards include a LinkedIn post at 1,600 cases and two videos by title. Beside it stand HAQQ's sanctions tracker, a vendor's monthly re-count that lists sanctions up to about $109,700, and Matthew Lee's international tracker of selected judicial quotations at Natural and Artificial Law, which measures no frequency, with a separate UK tracker whose inclusion is not proof (HAQQ; Natural and Artificial Law).", "note": "L_F(B) is realized as the field arm's expansion (objects.KO_B): the field alone, cards, own page and Wikipedia." }, "L_BA": { "words": 1678, "text": "AI Hallucination Cases is Damien Charlotin's database of legal decisions in which a court or tribunal finds or implies reliance on hallucinated material (MMRS record); its own page says it \"tracks legal decisions\" in cases where generative AI produced hallucinated content, \"typically fake citations, but also other types of AI-generated arguments\", and describes itself as \"the most comprehensive database of AI hallucination cases in law\". Charlotin wrote in November 2025 that he maintains it on his personal website (FAQ), and the MMRS record gives him as its operator by his own statement; HAQQ's tracker calls him a research fellow at HEC Paris, and Wikipedia's articles on artificial intelligence and on hallucination a legal data analysis lecturer at HEC Paris and Sciences Po. Its unit is the decision (MMRS record), which its page defines as any document \"where the use of AI, whether established or merely alleged, is addressed in more than a passing reference by the court or tribunal\"; it admits \"only cases where the court or tribunal has explicitly found (or implied) that a party relied on hallucinated content or material\", which the MMRS record states as reliance a court \"explicitly finds, or implies\", entering it in its self-description field and as the property that admission requires \"a court finding or implication of reliance on hallucinated material\". The judgment is the courts': \"I am not making that judgment, I let the courts and judges make or imply it, which is why the database is necessarily an undercount\" (FAQ); HAQQ restates the bar as a court that \"found, or clearly implied\" reliance, accusations alone not counting, and the FAQ holds that most hallucinations are \"very obvious\", a made-up case name or false quote one that \"cannot be a human mistake\". Its exceptions are stated twice, and the statements pull apart: its page \"also covers some judicial decisions where AI use was alleged but not confirmed\", \"a judgment call on my part\", while the FAQ says he refrains \"from adding rows about cases where hallucinations are only alleged\" and, in a February 2026 edit, makes an exception where the alleged hallucination comes from a judge and, absent an appeal decision or retraction, he is \"bound to make a judgment call\"; the MMRS record composes both as \"Mere allegations are excluded, with some judgment calls admitted.\" It states its boundary itself, that it \"does not track the (necessarily wider) universe of all fake citations or use of AI in court filings\" (its page; quoted by the MMRS record), and on the MMRS record's reading a fabricated citation enters only when a court catches and rules on it; HAQQ draws the same consequence for its reading of the count, \"a fake citation nobody checks never reaches the database, so 2,202 is a floor\". Each record gives the case, the court and jurisdiction, the date, the party using AI, the AI tool, the nature of the hallucination, the outcome or sanction, the monetary penalty, details and reports (its table; the MMRS record lists the court, date, party, tool where named, nature and outcome), the nature filed as Fabricated (1,829), False Quotes (595), Misrepresented (947) or Outdated Advice (38). Its outcomes can be filtered for monetary sanctions and disciplinary referrals, and a penalty recorded as 1 \"is a placeholder, not the real figure\", the amount being unknown (its page). Its page states the record's limits: AI tools are tracked \"only as disclosed or admitted by parties in court filings\", with 'Unidentified' and 'Implied' as the defaults; a named tool \"does not necessarily mean that tool was responsible\"; offending-document dates are parsed automatically and \"may occasionally be wrong\", and responses when caught are extracted automatically, \"indicative rather than exact\". Seeking to be exhaustive, it had identified 2,231 cases at its update of 28 October 2026 and calls itself \"a work in progress\" (its page; the MMRS record gives the same count as of the site's own update date), 1,514 of them from the USA, 238 from Canada, 112 from Australia and 82 from the UK, with 1,288 involving pro se litigants, 880 lawyers, 32 judges, 17 experts and 5 prosecutors (its filters; the MMRS record gives the pro se and lawyer counts). HAQQ, reading the database's update of 19 October 2026, gives 2,202 decisions, 1,501 from the United States, 1,267 involving pro se litigants and 872 lawyers. Charlotin started it in April 2025 while teaching Sciences Po Law School students on the \"Limits and Potentials\" of LLMs in law, when he looked for data on hallucinations and, finding none, decided to \"tally the cases myself\" (FAQ); Wikipedia dates its establishment to April 2025 and places its creation among the effects of law firms' use of generative AI for research; its period filter begins in the second quarter of 2023 (its page; MMRS record), and the MMRS record's census enters that filter as its founding. Before 2025, the FAQ says, \"we maybe had two or three cases a month\", and by February 2026 the daily average was five; Wikipedia's hallucination article gives over 1,300 instances a year after its start, a LinkedIn post four months before the capture 1,600 cases, and HAQQ's series about 200 in mid-2025, 719 in January 2026 and 2,202 on 19 October 2026; its page reads \"Updated daily\" and was last updated on 28 October 2026, and the MMRS record gives its cadence as continuous. The FAQ attributes the acceleration to judicial lag, the availability of LLMs and public knowledge of them, matching a profile of self-represented litigants or lawyers surprised by a tool's new AI component, and explains the share of US entries by the US legal data ecosystem, legal data in many European countries being \"hoarded by legal editors\". It divides the records into pure mistakes, \"the vast majority of cases\", and a \"(substantial) minority\" involving \"bad\" actors (FAQ). Cases come from referrals, scrapers and bots monitoring data sources, and keyword searches (FAQ), and readers are asked to send cases (its page); the MMRS record gives the intake as editor-curated with readers sending cases and the database growing by its editor's reading and readers' cases; the pipeline for new entries itself uses AI, \"though I am careful to check it does not hallucinate\" (FAQ). It is licensed CC BY 4.0, \"free to reuse, including commercially, with attribution\", to be cited as \"AI Hallucination Cases Database, Damien Charlotin\", and its cases download as CSV (its page; MMRS record). The FAQ gives its use as practitioners finding the cases relevant in their own jurisdiction, data analyses, and the cases as \"benchmarks to fix the issue\"; from it Charlotin built PelAIkan, \"an automated reference checker that also detects hallucinations\" (its page), and he expects no fix \"at the model stage\" and no reduction to zero \"given the existing paradigm\" (FAQ, a forecast). \"For years, we (me included) have cited without reading; but now the costs of that practice have become explicit\": hallucinations, the FAQ says, \"expose the epistemic hygiene the legal profession has long lacked\". The MMRS record reads it as \"the census's most exact record of citation fidelity\", recording attribution by the nature of the hallucination: \"Each entry is a machine composition that reached a court with sources it invented or misrepresented, and the court's decision is the reception that recorded it.\" It has been featured in news media \"and indeed in several decisions dealing with hallucinated material\", the LA Times among them in May 2025 (its page); Wikipedia cites it for counts of Canadian filings (at least 167 by 14 June 2026, with courtready.ca) and Israeli cases (at least 55 by August 2026) and, in its article on Mata v. Avianca, for that case's description as a landmark, and its article on artificial intelligence records judges' sanctions and bar warnings by 2026. On the Overview's video cards it appears at title grain: KARE 11's \"How common are AI hallucinations in court cases?\" and Miller Law Group's \"AI Hallucinations Got an Attorney Disbarred\". HAQQ, which discloses that it \"sells a legal AI product\", re-reads the live database monthly and takes off its page any number it cannot re-verify, and lists sanctions from revoked admissions to personal costs, the largest it can source about $109,700 (HAQQ); the MMRS record enters it as a candidate, \"a vendor tracker built on Charlotin's count\". Matthew Lee, a barrister at Doughty Street Chambers, keeps at Natural and Artificial Law an international tracker of \"Selected judicial quotations\" on AI hallucination cases and fabricated or false citations, first published 16 May 2025, which \"does not measure how often these issues occur\", and a separate UK tracker of case records, outcomes, quotations and sources in which \"inclusion is not proof of an AI hallucination or wrongdoing\"; neither page names Charlotin's database, which the MMRS record's candidate list enters as \"a UK companion to Charlotin's database\". The Capture Registry (2026-10-29) holds the UK tracker a separate aggregator and HAQQ's tracker, which gives the database's 2,202 decisions of 19 October (HAQQ), built on Charlotin's count, both added as candidates the census did not hold, and reads the cards as the database \"already aggregated in turn: a regional companion and a vendor tracker counting its cases\". The MMRS record keeps itself apart from the entity, each assertion with a source and a basis and the lifecycle apart from the observation that saw it, records the database as active on 28 October 2026 with the roles of monitoring and reception, and relates it to the composition layer's account seated in the Capture Registry. There (Capture Registry, 2026-10-29) the Overview at 'ai hallucination cases database' composes the database with its maker and at its size, Charlotin and 'over 2,200' being right against the 2,231 its page gives, as the field's primary record; it narrows the admission rule, which the page states as reliance the court \"has explicitly found (or implied)\", to 'explicitly', so that the implied cases are dropped, and lists private lawyers first where the page's filters give 1,288 pro se litigants and 880 lawyers.", "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": "Damien Charlotin — AI Hallucination Cases Database", "fetched": "page, HTTP 200 at 2026-10-29T05:06:47Z; raw sha256 68d8bdade886d77a…", "origin": "card" }, { "id": "B2", "card": "Natural and Artificial Law — AI Hallucination Cases Tracker", "fetched": "page, HTTP 200 at 2026-10-29T05:08:25Z; raw sha256 b4c8760c47ec182d…", "origin": "card" }, { "id": "B3", "card": "HAQQ — AI Hallucination Cases: The 2202-Case Sanctions Tracker - HAQQ", "fetched": "page, HTTP 200 at 2026-10-29T05:08:26Z; raw sha256 99fa409d712f0ae0…", "origin": "card" }, { "id": "B4", "card": "LinkedIn — George Hannah's Post", "fetched": "unreachable (not found by WebSearch); card title and snippet stand", "origin": "card" }, { "id": "B5", "card": "YouTube·KARE 11 — How common are AI hallucinations in court cases?", "fetched": "oEmbed, HTTP 200 at 2026-10-29T05:08:26Z; raw sha256 ac41df10addc1eb2…", "origin": "card" }, { "id": "B6", "card": "YouTube·Miller Law Group — AI Hallucinations Got an Attorney Disbarred", "fetched": "oEmbed, HTTP 200 at 2026-10-29T05:08:27Z; raw sha256 60504a8520783b95…", "origin": "card" }, { "id": "B7", "card": "Artificial Authority (Damien Charlotin) — Hallucinations Case Database - FAQ", "fetched": "page, HTTP 200 at 2026-10-29T05:08:27Z; raw sha256 93d0d01730155efc…", "origin": "own_page" }, { "id": "B8", "card": "Wikipedia — Artificial intelligence", "fetched": "raw wikitext and rendered page, HTTP 200 at 2026-10-29T05:09:20Z; raw sha256 8046fc18449a00eb…", "origin": "wikipedia" }, { "id": "B9", "card": "Wikipedia — Hallucination (artificial intelligence)", "fetched": "raw wikitext and rendered page, HTTP 200 at 2026-10-29T05:09:20Z; raw sha256 44c64b72efb8835a…", "origin": "wikipedia" }, { "id": "B10", "card": "Wikipedia — Mata v. Avianca, Inc.", "fetched": "raw wikitext and rendered page, HTTP 200 at 2026-10-29T05:09:21Z; raw sha256 e3a4791f5537949a…", "origin": "wikipedia" }, { "id": "W1", "card": "Wikipedia — no article ('AI Hallucination Cases', 'AI Hallucination Cases Database', 'Damien Charlotin')", "fetched": "HTTP 404 (raw and rendered), 2026-10-29 05:08 UTC; Special:Search 200", "origin": "wikipedia" } ], "field_claims": { "F1": { "source": "B1", "origin": "card", "locus": "opening ¶1", "quote": "This database tracks legal decisions … in cases where generative AI produced hallucinated content – typically fake citations, but also other types of AI-generated arguments.", "claim": "the database defined by its own page: it tracks legal decisions in cases where generative AI produced hallucinated content, typically fake citations, also other AI-generated arguments", "modality": "stipulation", "kind": "definition", "kernel": true, "const": true, "lineage": "what it is: legal decisions on AI hallucinations", "date": "2026-10-28 (the page's 'Last updated'; as fetched 2026-10-29)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 c645e242eb50b572…), whitespace-normalised; multi-part quote joined by ' … ', each part checked" }, "F2": { "source": "B1", "origin": "card", "locus": "opening ¶1, footnote 1 ¶1", "quote": "I.e., all documents where the use of AI, whether established or merely alleged, is addressed in more than a passing reference by the court or tribunal.", "claim": "a 'decision' is any document in which the court or tribunal addresses AI use, established or alleged, in more than a passing reference", "modality": "stipulation", "kind": "definition", "kernel": false, "const": true, "lineage": "the unit and the admission rule", "date": "2026-10-28 (the page's 'Last updated'; as fetched 2026-10-29)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 c645e242eb50b572…), whitespace-normalised" }, "F3": { "source": "B1", "origin": "card", "locus": "opening ¶1, footnote 1 ¶2", "quote": "Notably, this does not cover mere allegations of hallucinations, but only cases where the court or tribunal has explicitly found (or implied) that a party relied on hallucinated content or material.", "claim": "admission rule: not mere allegations of hallucinations, only cases where the court or tribunal explicitly found (or implied) reliance on hallucinated material", "modality": "stipulation", "kind": "definition", "kernel": true, "const": true, "lineage": "the unit and the admission rule", "date": "2026-10-28 (the page's 'Last updated'; as fetched 2026-10-29)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 c645e242eb50b572…), whitespace-normalised" }, "F4": { "source": "B1", "origin": "card", "locus": "opening ¶1, footnote 1 ¶3", "quote": "As an exception, the database also covers some judicial decisions where AI use was alleged but not confirmed. This is a judgment call on my part.", "claim": "an exception: some decisions where AI use was alleged but not confirmed, admitted as the maker's judgment call", "modality": "stipulation", "kind": "limit", "kernel": false, "const": false, "lineage": "exceptions: judgment calls", "date": "2026-10-28 (the page's 'Last updated'; as fetched 2026-10-29)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 c645e242eb50b572…), whitespace-normalised" }, "F5": { "source": "B1", "origin": "card", "locus": "opening ¶1", "quote": "It does not track the (necessarily wider) universe of all fake citations or use of AI in court filings.", "claim": "its boundary: it does not track the necessarily wider universe of all fake citations or AI use in court filings", "modality": "self-description", "kind": "limit", "kernel": true, "const": false, "lineage": "boundary: what it does not track", "date": "2026-10-28 (the page's 'Last updated'; as fetched 2026-10-29)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 c645e242eb50b572…), whitespace-normalised" }, "F6": { "source": "B1", "origin": "card", "locus": "opening ¶2", "quote": "While seeking to be exhaustive (2231 cases identified so far), it is a work in progress and will expand as new examples emerge.", "claim": "2,231 cases identified, seeking exhaustiveness; a work in progress", "modality": "self-description", "kind": "fact", "kernel": false, "const": false, "lineage": "count and composition", "date": "2026-10-28 (the page's 'Last updated'; as fetched 2026-10-29)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 c645e242eb50b572…), whitespace-normalised" }, "F7": { "source": "B1", "origin": "card", "locus": "opening ¶2", "quote": "This database has been featured in news media, and indeed in several decisions dealing with hallucinated material.", "claim": "featured in news media and in several decisions dealing with hallucinated material", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "reception: press, courts, encyclopedia, video", "date": "2026-10-28 (the page's 'Last updated'; as fetched 2026-10-29)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 c645e242eb50b572…), whitespace-normalised" }, "F8": { "source": "B1", "origin": "card", "locus": "opening ¶2, footnote 2", "quote": "Examples of media coverage include: … M. Hiltzik, AI 'hallucinations' are a growing problem for the legal profession (LA Times, 22 May 2025)", "claim": "media coverage, the first example the LA Times of 22 May 2025", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "reception: press, courts, encyclopedia, video", "date": "2026-10-28 (the page's 'Last updated'; as fetched 2026-10-29)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 c645e242eb50b572…), whitespace-normalised; multi-part quote joined by ' … ', each part checked" }, "F9": { "source": "B1", "origin": "card", "locus": "opening ¶3", "quote": "And if you know of a case that should be included, feel free to … contact me.", "claim": "readers may send cases to the maker", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "intake and pipeline", "date": "2026-10-28 (the page's 'Last updated'; as fetched 2026-10-29)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 c645e242eb50b572…), whitespace-normalised; multi-part quote joined by ' … ', each part checked" }, "F10": { "source": "B1", "origin": "card", "locus": "PelAIkan box", "quote": "Based on this database, I have developed an automated reference checker that also detects hallucinations: PelAIkan.", "claim": "a derived tool: PelAIkan, an automated reference checker built on the database", "modality": "documented", "kind": "function", "kernel": false, "const": false, "lineage": "purpose and uses", "date": "2026-10-28 (the page's 'Last updated'; as fetched 2026-10-29)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 c645e242eb50b572…), whitespace-normalised" }, "F11": { "source": "B1", "origin": "card", "locus": "download and update line", "quote": "Click to Download CSV … Last updated: 28 October 2026", "claim": "CSV download; last updated 28 October 2026", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "openness: licence and data", "date": "2026-10-28 (the page's 'Last updated'; as fetched 2026-10-29)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 c645e242eb50b572…), whitespace-normalised; multi-part quote joined by ' … ', each part checked" }, "F12": { "source": "B1", "origin": "card", "locus": "Licence box", "quote": "CC BY 4.0 — free to reuse, including commercially, with attribution. … Cite as: AI Hallucination Cases Database, Damien Charlotin,", "claim": "licensed CC BY 4.0, reuse including commercial with attribution; cited as 'AI Hallucination Cases Database, Damien Charlotin'", "modality": "stipulation", "kind": "fact", "kernel": false, "const": false, "lineage": "openness: licence and data", "date": "2026-10-28 (the page's 'Last updated'; as fetched 2026-10-29)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 c645e242eb50b572…), whitespace-normalised; multi-part quote joined by ' … ', each part checked" }, "F13": { "source": "B1", "origin": "card", "locus": "filters: Party", "quote": "Expert (17) … Judge (32) … Lawyer (880) … Prosecutor (5) … Pro Se Litigant (1288)", "claim": "by party: pro se litigants 1,288, lawyers 880, judges 32, experts 17, prosecutors 5", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "count and composition", "date": "2026-10-28 (the page's 'Last updated'; as fetched 2026-10-29)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 c645e242eb50b572…), whitespace-normalised; multi-part quote joined by ' … ', each part checked" }, "F14": { "source": "B1", "origin": "card", "locus": "filters: State", "quote": "Australia (112) … Canada (238) … UK (82) … USA (1514)", "claim": "by jurisdiction: USA 1,514, Canada 238, Australia 112, UK 82", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "count and composition", "date": "2026-10-28 (the page's 'Last updated'; as fetched 2026-10-29)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 c645e242eb50b572…), whitespace-normalised; multi-part quote joined by ' … ', each part checked" }, "F15": { "source": "B1", "origin": "card", "locus": "filters: Nature – Category", "quote": "Fabricated (1829) … False Quotes (595) … Misrepresented (947) … Outdated Advice (38)", "claim": "nature of hallucination in four categories: fabricated 1,829, false quotes 595, misrepresented 947, outdated advice 38", "modality": "documented", "kind": "definition", "kernel": false, "const": false, "lineage": "the record: fields and categories", "date": "2026-10-28 (the page's 'Last updated'; as fetched 2026-10-29)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 c645e242eb50b572…), whitespace-normalised; multi-part quote joined by ' … ', each part checked" }, "F16": { "source": "B1", "origin": "card", "locus": "filters: From period", "quote": "From period: Q2 2023", "claim": "the period filter begins in Q2 2023", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "maker and origin", "date": "2026-10-28 (the page's 'Last updated'; as fetched 2026-10-29)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 c645e242eb50b572…), whitespace-normalised" }, "F17": { "source": "B1", "origin": "card", "locus": "table header", "quote": "Court / Jurisdiction … Party Using AI … AI Tool … Nature of Hallucination … Outcome / Sanction … Monetary Penalty … Details … Report(s)", "claim": "each record: case, court and jurisdiction, date, party using AI, AI tool, nature of hallucination, outcome or sanction, monetary penalty, details, reports", "modality": "stipulation", "kind": "definition", "kernel": false, "const": false, "lineage": "the record: fields and categories", "date": "2026-10-28 (the page's 'Last updated'; as fetched 2026-10-29)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 c645e242eb50b572…), whitespace-normalised; multi-part quote joined by ' … ', each part checked" }, "F18": { "source": "B1", "origin": "card", "locus": "AI Tool column note (attribute text)", "quote": "This database tracks AI tools only as disclosed or admitted by parties in court filings. If AI use was acknowledged but no specific tool was named, the entry reads 'Unidentified.' Otherwise, the default value is 'Implied' when AI use is evident but not explicitly confirmed.", "claim": "AI tools recorded only as disclosed or admitted in filings; 'Unidentified' and 'Implied' as defaults", "modality": "stipulation", "kind": "limit", "kernel": false, "const": false, "lineage": "the record's stated limits", "date": "2026-10-28 (the page's 'Last updated'; as fetched 2026-10-29)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 c645e242eb50b572…), whitespace-normalised" }, "F19": { "source": "B1", "origin": "card", "locus": "AI Tool column note (attribute text)", "quote": "Note: The mention of a specific tool does not necessarily mean that tool was responsible for the hallucinations in question.", "claim": "a named tool is not necessarily the tool responsible", "modality": "self-description", "kind": "limit", "kernel": false, "const": false, "lineage": "the record's stated limits", "date": "2026-10-28 (the page's 'Last updated'; as fetched 2026-10-29)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 c645e242eb50b572…), whitespace-normalised" }, "F20": { "source": "B1", "origin": "card", "locus": "graph notes (attribute text)", "quote": "The date of the offending document is parsed automatically from court records and may occasionally be wrong. … These responses are extracted automatically from the decisions and simplified into set categories, so treat them as indicative rather than exact.", "claim": "automatically parsed dates may be wrong; responses when caught are extracted automatically and indicative rather than exact", "modality": "self-description", "kind": "limit", "kernel": false, "const": false, "lineage": "the record's stated limits", "date": "2026-10-28 (the page's 'Last updated'; as fetched 2026-10-29)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 c645e242eb50b572…), whitespace-normalised; multi-part quote joined by ' … ', each part checked" }, "F21": { "source": "B1", "origin": "card", "locus": "Monetary sanction note (attribute text)", "quote": "A monetary penalty recorded as 1 is a placeholder, not the real figure: the court imposed a monetary sanction but the amount is unknown to us.", "claim": "a penalty of 1 is a placeholder for an unknown amount", "modality": "stipulation", "kind": "definition", "kernel": false, "const": false, "lineage": "the record: fields and categories", "date": "2026-10-28 (the page's 'Last updated'; as fetched 2026-10-29)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 c645e242eb50b572…), whitespace-normalised" }, "F22": { "source": "B1", "origin": "card", "locus": "meta description (attribute text)", "quote": "The most comprehensive database of AI hallucination cases in law: legal decisions from courts worldwide, searchable by country, party, AI tool, and outcome. Updated daily.", "claim": "its self-description: the most comprehensive database of AI hallucination cases in law, worldwide, searchable, updated daily", "modality": "self-description", "kind": "definition", "kernel": false, "const": false, "lineage": "what it is: legal decisions on AI hallucinations", "date": "2026-10-28 (the page's 'Last updated'; as fetched 2026-10-29)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 c645e242eb50b572…), whitespace-normalised" }, "F23": { "source": "B1", "origin": "card", "locus": "filters: outcome", "quote": "Monetary sanction … Disciplinary Referral", "claim": "outcome filters: monetary sanction, disciplinary referral", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "the record: fields and categories", "date": "2026-10-28 (the page's 'Last updated'; as fetched 2026-10-29)", "quote_ok": true, "checked_against": "saved text B1.txt (sha256 c645e242eb50b572…), whitespace-normalised; multi-part quote joined by ' … ', each part checked" }, "F24": { "source": "B2", "origin": "card", "locus": "UK box", "quote": "The detailed UK AI Hallucination Cases Tracker … For UK AI hallucination cases, the dedicated UK tracker brings together case records, recorded outcomes, judicial quotations and sources, alongside figures and charts drawn from my fuller research.", "claim": "Natural and Artificial Law's separate UK tracker: case records, outcomes, judicial quotations, sources, figures and charts", "modality": "documented", "kind": "definition", "kernel": false, "const": false, "lineage": "neighbour: Natural and Artificial Law's trackers", "date": "2025-05-16 (first published; revised 2026-10-02)", "quote_ok": true, "checked_against": "saved text B2.txt (sha256 bd7198c1f5306d58…), whitespace-normalised; multi-part quote joined by ' … ', each part checked" }, "F25": { "source": "B2", "origin": "card", "locus": "UK box", "quote": "It includes confirmed, suspected and disputed issues: inclusion is not proof of an AI hallucination or wrongdoing.", "claim": "the UK tracker includes confirmed, suspected and disputed issues; inclusion is not proof", "modality": "stipulation", "kind": "limit", "kernel": false, "const": false, "lineage": "neighbour: Natural and Artificial Law's trackers", "date": "2025-05-16 (first published; revised 2026-10-02)", "quote_ok": true, "checked_against": "saved text B2.txt (sha256 bd7198c1f5306d58…), whitespace-normalised" }, "F26": { "source": "B2", "origin": "card", "locus": "international page head", "quote": "International AI hallucination cases tracker … Selected judicial quotations on AI hallucination cases and fabricated/false citations from judgments and decisions across jurisdictions. … By Matthew Lee · Barrister, Doughty Street Chambers … First published: 16 May 2025", "claim": "the card's page: Matthew Lee's international tracker of selected judicial quotations on AI hallucination cases, first published 16 May 2025", "modality": "stipulation", "kind": "definition", "kernel": true, "const": false, "lineage": "neighbour: Natural and Artificial Law's trackers", "date": "2025-05-16 (first published; revised 2026-10-02)", "quote_ok": true, "checked_against": "saved text B2.txt (sha256 bd7198c1f5306d58…), whitespace-normalised; multi-part quote joined by ' … ', each part checked" }, "F27": { "source": "B2", "origin": "card", "locus": "The reason for this collection", "quote": "The collection is selective and does not measure how often these issues occur or establish a worldwide consensus.", "claim": "the collection is selective; it measures no frequency and establishes no consensus", "modality": "self-description", "kind": "limit", "kernel": true, "const": false, "lineage": "neighbour: Natural and Artificial Law's trackers", "date": "2025-05-16 (first published; revised 2026-10-02)", "quote_ok": true, "checked_against": "saved text B2.txt (sha256 bd7198c1f5306d58…), whitespace-normalised" }, "F28": { "source": "B3", "origin": "card", "locus": "lede ('In short')", "quote": "In short: Damien Charlotin's AI Hallucination Cases database lists 2,202 court and tribunal decisions involving AI-fabricated citations or content, as of its 19 October 2026 update.", "claim": "HAQQ: the database lists 2,202 decisions as of its 19 October 2026 update", "modality": "attributed", "kind": "fact", "kernel": true, "const": false, "lineage": "count and composition", "date": "2026-05-22 (published; updated 2026-10-20)", "quote_ok": true, "checked_against": "saved text B3.txt (sha256 7a350124b3e5de72…), whitespace-normalised" }, "F29": { "source": "B3", "origin": "card", "locus": "How many AI hallucination cases are there?", "quote": "The database is maintained by Damien Charlotin, a research fellow at HEC Paris. Its inclusion bar is strict: a decision is listed only where a court or tribunal found, or clearly implied, that a party relied on hallucinated material. Accusations alone do not count.", "claim": "HAQQ: maintained by Charlotin, a research fellow at HEC Paris; inclusion only where a court found or clearly implied reliance", "modality": "attributed", "kind": "attribution", "kernel": false, "const": false, "lineage": "the unit and the admission rule", "date": "2026-05-22 (published; updated 2026-10-20)", "quote_ok": true, "checked_against": "saved text B3.txt (sha256 7a350124b3e5de72…), whitespace-normalised" }, "F30": { "source": "B3", "origin": "card", "locus": "How many AI hallucination cases are there? (caveats)", "quote": "First, it counts only what a judge caught and wrote up; a fake citation nobody checks never reaches the database, so 2,202 is a floor.", "claim": "HAQQ: it counts only what a judge caught; the count is a floor", "modality": "interpretation", "kind": "limit", "kernel": true, "const": false, "lineage": "boundary: what it does not track", "date": "2026-05-22 (published; updated 2026-10-20)", "quote_ok": true, "checked_against": "saved text B3.txt (sha256 7a350124b3e5de72…), whitespace-normalised" }, "F31": { "source": "B3", "origin": "card", "locus": "Key facts", "quote": "By country: United States 1,501; Canada 234; Australia 112; United Kingdom 76; … By filer: pro se litigants 1,267; lawyers 872; judges 32; experts 17 (Charlotin database party filter, 19 October 2026).", "claim": "HAQQ's reading of the filters on 19 October 2026: US 1,501; pro se 1,267, lawyers 872", "modality": "attributed", "kind": "fact", "kernel": false, "const": false, "lineage": "count and composition", "date": "2026-05-22 (published; updated 2026-10-20)", "quote_ok": true, "checked_against": "saved text B3.txt (sha256 7a350124b3e5de72…), whitespace-normalised; multi-part quote joined by ' … ', each part checked" }, "F32": { "source": "B3", "origin": "card", "locus": "Key facts", "quote": "Growth: about 200 cases in mid-2025, 719 in January 2026 (Scientific American, 2026), 1,227 in early April 2026 (PlatinumIDS, April 2026), 1,458 on 22 May, 1,598 on 9 June, 2,046 on 21 September and 2,202 on 19 October 2026 (HAQQ checks of the live database).", "claim": "HAQQ's growth series, about 200 in mid-2025 to 2,202 on 19 October 2026", "modality": "attributed", "kind": "fact", "kernel": false, "const": false, "lineage": "growth", "date": "2026-05-22 (published; updated 2026-10-20)", "quote_ok": true, "checked_against": "saved text B3.txt (sha256 7a350124b3e5de72…), whitespace-normalised" }, "F33": { "source": "B3", "origin": "card", "locus": "How have AI sanctions escalated?", "quote": "The 2025 and 2026 decisions add revoked admissions, regulator referrals, a contempt finding, a canceled trial, district suspensions and personal costs against the lawyer who signed without checking. … Largest single-matter penalty we can source: about $109,700 in sanctions and fees in Couvrette v. Wisnovsky, D. Oregon (ABA Journal, 2026).", "claim": "HAQQ: sanctions beyond fines (revoked admissions, referrals, suspensions, personal costs); largest sourced penalty about $109,700", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "neighbour: HAQQ's sanctions tracker", "date": "2026-05-22 (published; updated 2026-10-20)", "quote_ok": true, "checked_against": "saved text B3.txt (sha256 7a350124b3e5de72…), whitespace-normalised; multi-part quote joined by ' … ', each part checked" }, "F34": { "source": "B3", "origin": "card", "locus": "How is this tracker kept current?", "quote": "Once a month we re-read the live database and primary reporting and check four things: the total count, the largest sourced penalty, new landmark decisions (appellate, cross-border or a new kind of sanction) and the implied daily rate. A number we cannot re-verify comes off the page.", "claim": "HAQQ's method: a monthly re-read of the live database; unverifiable numbers removed", "modality": "self-description", "kind": "function", "kernel": false, "const": false, "lineage": "neighbour: HAQQ's sanctions tracker", "date": "2026-05-22 (published; updated 2026-10-20)", "quote_ok": true, "checked_against": "saved text B3.txt (sha256 7a350124b3e5de72…), whitespace-normalised" }, "F35": { "source": "B3", "origin": "card", "locus": "Why do lawyers keep filing fake AI citations?", "quote": "(HAQQ ran that benchmark and sells a legal AI product, so weigh our numbers accordingly)", "claim": "HAQQ discloses that it sells a legal AI product", "modality": "self-description", "kind": "limit", "kernel": false, "const": false, "lineage": "neighbour: HAQQ's sanctions tracker", "date": "2026-05-22 (published; updated 2026-10-20)", "quote_ok": true, "checked_against": "saved text B3.txt (sha256 7a350124b3e5de72…), whitespace-normalised" }, "F36": { "source": "B4", "origin": "card", "locus": "card snippet (LinkedIn; post not found)", "quote": "George Hannah 4mo There is now a database tracking every AI hallucination to surface in court. It has logged 1,600 cases so far.", "claim": "a LinkedIn post (four months before the capture): a database tracking every AI hallucination to surface in court, 1,600 cases", "modality": "attributed", "kind": "fact", "kernel": true, "const": false, "lineage": "reception: press, courts, encyclopedia, video", "date": "about 2026-06 ('4mo' on the card, 2026-10-29)", "quote_ok": true, "checked_against": "the card's title/snippet in the registry's cite_list" }, "F37": { "source": "B5", "origin": "card", "locus": "oEmbed title (YouTube, KARE 11, 1:15)", "quote": "How common are AI hallucinations in court cases?", "claim": "a KARE 11 video, 'How common are AI hallucinations in court cases?' (title only)", "modality": "documented", "kind": "fact", "kernel": true, "const": false, "lineage": "reception: press, courts, encyclopedia, video", "date": "— (title only)", "quote_ok": true, "checked_against": "saved text B5.txt (sha256 22256d67ebcd30fb…), whitespace-normalised" }, "F38": { "source": "B6", "origin": "card", "locus": "oEmbed title (YouTube, Miller Law Group, 0:53)", "quote": "AI Hallucinations Got an Attorney Disbarred", "claim": "a Miller Law Group video, 'AI Hallucinations Got an Attorney Disbarred' (title only)", "modality": "documented", "kind": "fact", "kernel": true, "const": false, "lineage": "reception: press, courts, encyclopedia, video", "date": "— (title only)", "quote_ok": true, "checked_against": "saved text B6.txt (sha256 2b0789bf3bdb1b25…), whitespace-normalised" }, "F39": { "source": "B7", "origin": "own_page", "locus": "¶1", "quote": "As some of you know, I am maintaining an AI Hallucinations Cases database on my personal website.", "claim": "the maker: Charlotin maintains the database on his personal website", "modality": "documented", "kind": "attribution", "kernel": false, "const": false, "lineage": "maker and origin", "date": "2025-11-14 (edits of February 2026)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 9e2ad64a4347ae04…), whitespace-normalised" }, "F40": { "source": "B7", "origin": "own_page", "locus": "When did you start the database, and why ?", "quote": "April 2025. I was teaching this course to sorry Sciences Po Law School students, discussing in particular the “Limits and Potentials” of LLMs in the legal domain, when we started touching on the phenomenon of hallucinations. … I looked out for actual data here. Since there was none, … I figured, “hey, let’s tally the cases myself”.", "claim": "started April 2025 in a Sciences Po course on LLMs in law; no data existed, so he tallied the cases himself", "modality": "documented", "kind": "genealogy", "kernel": true, "const": false, "lineage": "maker and origin", "date": "2025-11-14 (edits of February 2026)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 9e2ad64a4347ae04…), whitespace-normalised; multi-part quote joined by ' … ', each part checked" }, "F41": { "source": "B7", "origin": "own_page", "locus": "What do you mean by “surge” ?", "quote": "I mean that pre-2025, we maybe had two or three cases a month; now that’s below the daily average. … daily average is now five per day.", "claim": "growth: two or three a month before 2025; five a day by February 2026", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "growth", "date": "2025-11-14 (edits of February 2026)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 9e2ad64a4347ae04…), whitespace-normalised; multi-part quote joined by ' … ', each part checked" }, "F42": { "source": "B7", "origin": "own_page", "locus": "Why the acceleration then ?", "quote": "A combination of factors, including the lag in judicial times … This matches the profile of most cases: either self-represented litigants, or lawyers who are surprised that an existing tool suddenly has an AI component that hallucinates.", "claim": "causes: judicial lag, LLM availability, public knowledge; most cases self-represented litigants or lawyers surprised by an AI component", "modality": "interpretation", "kind": "interpretation", "kernel": false, "const": false, "lineage": "causes and coverage: the maker's reading", "date": "2025-11-14 (edits of February 2026)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 9e2ad64a4347ae04…), whitespace-normalised; multi-part quote joined by ' … ', each part checked" }, "F43": { "source": "B7", "origin": "own_page", "locus": "How do you find the cases ?", "quote": "A mix of referrals from benevolent randos on the internet … use of dedicated scrapers and bots to automatically monitor some data sources … and good old database searches with, wait for it, keywords.", "claim": "intake: referrals, scrapers and bots, keyword searches", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "intake and pipeline", "date": "2025-11-14 (edits of February 2026)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 9e2ad64a4347ae04…), whitespace-normalised; multi-part quote joined by ' … ', each part checked" }, "F44": { "source": "B7", "origin": "own_page", "locus": "But then, how do you know a case actually involves an hallucination ?", "quote": "That’s the point: I am not making that judgment, I let the courts and judges make or imply it, which is why the database is necessarily an undercount.", "claim": "the courts make or imply the judgment, not the maker; hence necessarily an undercount", "modality": "stipulation", "kind": "limit", "kernel": true, "const": false, "lineage": "the unit and the admission rule", "date": "2025-11-14 (edits of February 2026)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 9e2ad64a4347ae04…), whitespace-normalised" }, "F45": { "source": "B7", "origin": "own_page", "locus": "But then, how do you know a case actually involves an hallucination ?", "quote": "It’s also why I refrain from adding rows about cases where hallucinations are only alleged", "claim": "he refrains from adding cases where hallucinations are only alleged", "modality": "stipulation", "kind": "limit", "kernel": false, "const": false, "lineage": "exceptions: judgment calls", "date": "2025-11-14 (edits of February 2026)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 9e2ad64a4347ae04…), whitespace-normalised" }, "F46": { "source": "B7", "origin": "own_page", "locus": "But then, how do you know a case actually involves an hallucination ?", "quote": "by their nature, most hallucinations are very obvious, making up a case name or a false quote is not, cannot be a human mistake", "claim": "most hallucinations are obvious; a made-up case name or false quote cannot be a human mistake", "modality": "interpretation", "kind": "interpretation", "kernel": false, "const": false, "lineage": "the unit and the admission rule", "date": "2025-11-14 (edits of February 2026)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 9e2ad64a4347ae04…), whitespace-normalised" }, "F47": { "source": "B7", "origin": "own_page", "locus": "But then, how do you know … ? (edit, Feb 2026)", "quote": "[Edit, Feb 2026: I should add an exception here, in cases where the alleged hallucination comes from a judge: absent an appeal court decision or an official retractation, I am bound to make a judgment call.]", "claim": "an exception for hallucinations alleged of a judge: a judgment call absent an appeal decision or retraction", "modality": "stipulation", "kind": "limit", "kernel": false, "const": false, "lineage": "exceptions: judgment calls", "date": "2026-02 (edit)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 9e2ad64a4347ae04…), whitespace-normalised" }, "F48": { "source": "B7", "origin": "own_page", "locus": "Why so many entries come from the USA ?", "quote": "‘Murica is blessed - and I mean it - with an excellent legal data ecosystem. … In many countries, especially European countries, legal data - though supposedly public - is hoarded by legal editors", "claim": "coverage: the US legal data ecosystem; European legal data hoarded by editors", "modality": "interpretation", "kind": "interpretation", "kernel": false, "const": false, "lineage": "causes and coverage: the maker's reading", "date": "2025-11-14 (edits of February 2026)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 9e2ad64a4347ae04…), whitespace-normalised; multi-part quote joined by ' … ', each part checked" }, "F49": { "source": "B7", "origin": "own_page", "locus": "Any other trends in the data ?", "quote": "one rather evident divisio stands between pure mistakes … the vast majority of cases, and the (substantial) minority of records that involve, for lack of a better word, “bad” actors.", "claim": "pure mistakes the vast majority; a substantial minority of 'bad' actors", "modality": "interpretation", "kind": "interpretation", "kernel": false, "const": false, "lineage": "causes and coverage: the maker's reading", "date": "2025-11-14 (edits of February 2026)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 9e2ad64a4347ae04…), whitespace-normalised; multi-part quote joined by ' … ', each part checked" }, "F50": { "source": "B7", "origin": "own_page", "locus": "When do you intend to stop ?", "quote": "I currently have a rather efficient pipeline to process new entries - which in fact involve the use of AI, though I am careful to check it does not hallucinate.", "claim": "the pipeline for new entries uses AI, checked for hallucination", "modality": "self-description", "kind": "fact", "kernel": false, "const": false, "lineage": "intake and pipeline", "date": "2025-11-14 (edits of February 2026)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 9e2ad64a4347ae04…), whitespace-normalised" }, "F51": { "source": "B7", "origin": "own_page", "locus": "What’s the point of the database ultimately ?", "quote": "Intrisic value: practitioners use it to find out what cases are relevant in their own jurisdiction. … And then all these hallucinated cases can serve as benchmarks to fix the issue.", "claim": "uses: practitioners' jurisdiction research, data analyses, benchmarks", "modality": "stipulation", "kind": "function", "kernel": false, "const": false, "lineage": "purpose and uses", "date": "2025-11-14 (edits of February 2026)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 9e2ad64a4347ae04…), whitespace-normalised; multi-part quote joined by ' … ', each part checked" }, "F52": { "source": "B7", "origin": "own_page", "locus": "Because you expect this to be fixed ?", "quote": "likely not at the model stage, no … I don’t think we can ever reduce it to zero given the existing paradigm.", "claim": "no fix expected at the model stage; not reducible to zero in the existing paradigm", "modality": "hypothesis", "kind": "prediction", "kernel": false, "const": false, "lineage": "purpose and uses", "date": "2025-11-14 (edits of February 2026)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 9e2ad64a4347ae04…), whitespace-normalised; multi-part quote joined by ' … ', each part checked" }, "F53": { "source": "B7", "origin": "own_page", "locus": "Anything else ?", "quote": "For years, we (me included) have cited without reading; but now the costs of that practice have become explicit. In other words, hallucinations expose the epistemic hygiene the legal profession has long lacked, and that is precisely why they deserve to be studied.", "claim": "hallucinations expose the profession's citing without reading", "modality": "interpretation", "kind": "interpretation", "kernel": false, "const": false, "lineage": "citation fidelity", "date": "2025-11-14 (edits of February 2026)", "quote_ok": true, "checked_against": "saved text B7.txt (sha256 9e2ad64a4347ae04…), whitespace-normalised" }, "F54": { "source": "B8", "origin": "wikipedia", "locus": "Applications / Other industry-specific tasks", "quote": "The use of generative AI by law firms for legal research resulted in the creation of the global \"AI Hallucination Cases\" database, in April 2025, established by HEC Paris and Sciences Po legal data analysis lecturer Damien Charlotin.", "claim": "Wikipedia (Artificial intelligence): created April 2025 by Charlotin, HEC Paris and Sciences Po lecturer, as a result of law firms' use of generative AI", "modality": "documented", "kind": "genealogy", "kernel": true, "const": false, "lineage": "maker and origin", "date": "2026-10-28 (revision listed by Special:Search)", "quote_ok": true, "checked_against": "saved text B8.txt (sha256 00d1de94af8ae826…), whitespace-normalised" }, "F55": { "source": "B8", "origin": "wikipedia", "locus": "Applications / Other industry-specific tasks", "quote": "By 2026, judges had issued sanctions and bar associations had issued warnings due to attorney submissions to the courts containing fabricated case law citations hallucinated by AI tools.", "claim": "Wikipedia (Artificial intelligence): by 2026 sanctions and bar warnings over AI-hallucinated citations", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "reception: press, courts, encyclopedia, video", "date": "2026-10-28 (revision listed by Special:Search)", "quote_ok": true, "checked_against": "saved text B8.txt (sha256 00d1de94af8ae826…), whitespace-normalised" }, "F56": { "source": "B9", "origin": "wikipedia", "locus": "In legal filings / United States ¶1", "quote": "A database of incidents, \"AI Hallucination Cases\", was established in April 2025 by HEC Paris and Sciences Po legal data analysis lecturer Damien Charlotin; one year later, it listed over 1300 instances of hallucinations in legal decisions.", "claim": "Wikipedia (Hallucination): established April 2025 by Charlotin; over 1,300 instances a year later", "modality": "documented", "kind": "genealogy", "kernel": true, "const": false, "lineage": "maker and origin", "date": "2026-10-28 (revision listed by Special:Search)", "quote_ok": true, "checked_against": "saved text B9.txt (sha256 d0bb8da24c35c8ef…), whitespace-normalised" }, "F57": { "source": "B9", "origin": "wikipedia", "locus": "In legal filings / Canada", "quote": "As of 14 June 2026, Canadian courts and tribunals have received at least 167 filings containing AI-hallucinated, fictitious citations, across 51 courts and tribunals.", "claim": "Wikipedia (Hallucination), sourced to courtready.ca and the database's Canada filter: at least 167 Canadian filings by 14 June 2026", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "reception: press, courts, encyclopedia, video", "date": "2026-10-28 (revision listed by Special:Search)", "quote_ok": true, "checked_against": "saved text B9.txt (sha256 d0bb8da24c35c8ef…), whitespace-normalised" }, "F58": { "source": "B9", "origin": "wikipedia", "locus": "In legal filings / Israel", "quote": "As of August 2026, Israeli courts and tribunals had identified at least 55 cases in which parties submitted documents containing fictitious citations or otherwise AI-hallucinated arguments.", "claim": "Wikipedia (Hallucination), sourced to the database's Israel filter: at least 55 Israeli cases by August 2026", "modality": "documented", "kind": "fact", "kernel": false, "const": false, "lineage": "reception: press, courts, encyclopedia, video", "date": "2026-10-28 (revision listed by Special:Search)", "quote_ok": true, "checked_against": "saved text B9.txt (sha256 d0bb8da24c35c8ef…), whitespace-normalised" }, "F59": { "source": "B10", "origin": "wikipedia", "locus": "Impact ¶3", "quote": "Due to the continued usage of GAI in the practice of law, Mata has been described as a landmark case by legal professionals,", "claim": "Wikipedia (Mata v. Avianca), sourced to the database among others: Mata described as a landmark case", "modality": "documented", "kind": "fact", "kernel": true, "const": false, "lineage": "reception: press, courts, encyclopedia, video", "date": "2026-05-19 (revision listed by Special:Search)", "quote_ok": true, "checked_against": "saved text B10.txt (sha256 0639347792d8bc82…), whitespace-normalised" } }, "field_sources": { "B1": "Damien Charlotin — AI Hallucination Cases Database", "B2": "Natural and Artificial Law — AI Hallucination Cases Tracker", "B3": "HAQQ — AI Hallucination Cases: The 2202-Case Sanctions Tracker - HAQQ", "B4": "LinkedIn — George Hannah's Post", "B5": "YouTube·KARE 11 — How common are AI hallucinations in court cases?", "B6": "YouTube·Miller Law Group — AI Hallucinations Got an Attorney Disbarred", "B7": "Artificial Authority (Damien Charlotin) — Hallucinations Case Database - FAQ", "B8": "Wikipedia — Artificial intelligence", "B9": "Wikipedia — Hallucination (artificial intelligence)", "B10": "Wikipedia — Mata v. Avianca, Inc." }, "delta": { "T_vs_LB": "At 'ai hallucination cases database' (T1, six cards: B1–B6) the composition gives the database by its page and its neighbours' pages. Composed from its cards: the maker (B1's 'Cite as … Damien Charlotin', F12; HAQQ's 'maintained by Damien Charlotin', F29), the size ('over 2,200': B1's 2,231, F6, or HAQQ's 2,202, F28), the US majority (B1 F14; HAQQ F31), the party categories (B1 F13), the nature categories at a coarser grain (B1 F15: 'fake legal citations, fabricated quotes, and non-existent case law' for Fabricated and False Quotes), the access link (B1) and the UK pointer (B2 F24). Force raised: 'The primary public database' composes the page's self-description, 'The most comprehensive database of AI hallucination cases in law' (F22, attribute text of B1), as a ranking, without attribution (coarse grain). Limit dropped: 'where generative AI explicitly produced hallucinated content' drops the page's '(or implied)' (F3) and with it the implied cases, and drops the exception for some decisions where AI use was alleged but not confirmed (F4) (CAP-1-02); 'most commonly fake legal citations, fabricated quotes, and non-existent case law' drops the Misrepresented (947) and Outdated Advice (38) categories (F15) and 'other types of AI-generated arguments' (F1); 'over 2,200 documented decisions' drops the page's boundary, that it 'does not track the (necessarily wider) universe of all fake citations or use of AI in court filings' (F5), and HAQQ's 'floor' (F30). Substituted reading: the parties are listed with 'private lawyers' first, where the filters give 1,288 pro se litigants against 880 lawyers (F13; CAP-1-04), and 'private lawyers' is not one of the page's categories (Lawyer, Governement Lawyer, Federal Defender); 'a companion resource' for UK records names a relation the card's page does not state: B2 is Matthew Lee's international tracker of selected judicial quotations (F26), which points to a separate UK tracker (F24) and does not name Charlotin or his database (MMRS-24 defect; CAP-1-03). Partly absent from the readable field: T1-5's 'mandatory verification rules' and 'disbarment'; the page's outcome filters are monetary sanction and disciplinary referral (F23), HAQQ lists revoked admissions, referrals, suspensions and personal costs (F33), and 'disbarred' appears only in card B6's title (F38, title only): if T drew on it, it composed a video title without attribution, which the field cannot settle. Available, not composed: the decision as any document addressing AI use beyond a passing reference (F2); the courts' judgment and the undercount (F44 is the FAQ's, not a card; on the cards HAQQ's 'counts only what a judge caught', F30); the AI-tool and extraction limits (F18–F20) and the penalty placeholder (F21); the licence, CC BY 4.0 with attribution, and the CSV (F11, F12); the reader intake (F9); PelAIkan (F10); media and court citations of the database (F7, F8); HAQQ's method and its disclosure that it sells a legal AI product (F34, F35); Natural and Artificial Law's own limits, selective and not measuring frequency, inclusion not proof (F25, F27); the LinkedIn post's 1,600 (F36). Content replaced by use: the closer (T1-8). Neither the FAQ (B7: the start in April 2025, the intake, the pipeline, the reading of the cases, citing without reading) nor Wikipedia (B8–B10) is a card; they are admitted beside the cards by the operator's ruling and were not surfaced by T.", "T_vs_LB_reading": "v0.9 §0.5: the first diff is already ontological, and here it is small on the entity and large on its limits. T and the field agree on the entity (Charlotin's database of court decisions on AI hallucinations, at its size); what T keeps is the page's self-description, its counts and its categories, and what it drops is the page's own conditions on what the count is: '(or implied)', the alleged-but-unconfirmed exception, the boundary of the wider universe, and the record's stated limits. It raises the page's self-description to a ranking and composes a neighbour as a companion. The row records this as observed and claims nothing about the composer's internals (§0.5).", "LB_vs_LBA": "Admission adds, in one claim space with the field's: the record's reading of what the selection leaves out, that a fabricated citation enters only when a court catches and rules on it (MMRS-10), beside the page's boundary (F5) and HAQQ's floor (F30); the archive's sense the field does not state, the database as the census's most exact record of citation fidelity, each entry 'a machine composition that reached a court with sources it invented or misrepresented', the court's decision the reception that recorded it (MMRS-11, MMRS-12), with attribution recorded by the nature of the hallucination (MMRS-22); the record kept apart from the entity, each field with source and basis, lifecycle apart from the observation (MMRS-26, MMRS-14) and the relation to the composition layer's account (MMRS-23); the neighbours as candidates (MMRS-24, MMRS-25) and the Capture Registry's observation of them (CAP-1-03, CAP-1-06); and the composition layer's treatment of the entity at its address (CAP-1-01, CAP-1-02, CAP-1-04, CAP-1-05). Coherence (P_coh, composed each with its source): the MMRS census's 'founded' (the period filter, Q2 2023, MMRS-20) against April 2025 (F40, F54, F56); 'a UK companion to Charlotin's database' (MMRS-24) against Natural and Artificial Law's pages, which do not name it (F24, F26); the MMRS record's 'Mere allegations are excluded, with some judgment calls admitted' (MMRS-04) beside the maker's two statements (F4; F45, F47). Restated lineages (§3.6, the own page keeping the earlier date): MMRS-01 and MMRS-16 with F1; MMRS-02, MMRS-03, MMRS-21 with F2, F3; MMRS-05 with F17, F15; MMRS-06 with F13; MMRS-07 with F16; MMRS-08, MMRS-17 with F9, F43; MMRS-09 with F5; MMRS-13 with F12, F39; MMRS-15 with F11, F12; MMRS-18 with F6; MMRS-19 with F22. The measure is in row.measure.", "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", "F33", "F34", "F35", "F36", "F37", "F38" ], "T1_composed": [ "F1", "F3", "F6", "F12", "F13", "F14", "F15", "F22", "F24", "F28", "F29", "F31" ], "T1_composed_count": "12 of 38 card claims", "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": "`MMRS-01`", "source": "MMRS record", "M_src": "stipulation", "sense": "what it is: legal decisions on AI hallucinations", "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 F1; the own page keeps the earlier date)", "resolver": "a later capture at 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K2", "claim": "`MMRS-02`", "source": "MMRS record", "M_src": "stipulation", "sense": "the unit and the admission rule", "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 F2, F3)", "resolver": "a later capture at 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K3", "claim": "`MMRS-03`", "source": "MMRS record", "M_src": "attributed", "sense": "the unit and the admission rule", "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 F3)", "resolver": "a later capture at 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K4", "claim": "`MMRS-04`", "source": "MMRS record", "M_src": "attributed", "sense": "exceptions: judgment calls", "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 (F4 against F45, F47)", "resolver": "a later capture at 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K5", "claim": "`MMRS-05`", "source": "MMRS record", "M_src": "attributed", "sense": "the record: fields and categories", "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 F17)", "resolver": "a later capture at 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K6", "claim": "`MMRS-06`", "source": "MMRS record", "M_src": "documented", "sense": "count and composition", "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 F13)", "resolver": "a later capture at 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K7", "claim": "`MMRS-07`", "source": "MMRS record", "M_src": "documented", "sense": "maker and origin", "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 F16)", "resolver": "a later capture at 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K8", "claim": "`MMRS-08`", "source": "MMRS record", "M_src": "documented", "sense": "intake and 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": "none (one lineage with F9, F43)", "resolver": "a later capture at 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K9", "claim": "`MMRS-09`", "source": "MMRS record", "M_src": "attributed", "sense": "boundary: what it does not track", "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 (quotes F5)", "resolver": "a later capture at 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K10", "claim": "`MMRS-10`", "source": "MMRS record", "M_src": "interpretation", "sense": "boundary: what it does not track", "qualifiers carried": "MMRS-09", "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 (F5, F30, F44)", "resolver": "a later capture at 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K11", "claim": "`MMRS-11`", "source": "MMRS record", "M_src": "interpretation", "sense": "citation fidelity", "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", "resolver": "a later capture at 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K12", "claim": "`MMRS-12`", "source": "MMRS record", "M_src": "interpretation", "sense": "citation fidelity", "qualifiers carried": "MMRS-11, MMRS-22", "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", "resolver": "a later capture at 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K13", "claim": "`MMRS-13`", "source": "MMRS record", "M_src": "documented", "sense": "maker and origin", "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 F12, F39)", "resolver": "a later capture at 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K14", "claim": "`MMRS-14`", "source": "MMRS record", "M_src": "documented", "sense": "the record kept apart from the entity", "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", "resolver": "a later capture at 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K15", "claim": "`MMRS-15`", "source": "MMRS record", "M_src": "documented", "sense": "openness: licence and 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": "none (one lineage with F11, F12)", "resolver": "a later capture at 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K16", "claim": "`MMRS-16`", "source": "MMRS record", "M_src": "attributed", "sense": "what it is: legal decisions on AI hallucinations", "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 (compresses F1–F3)", "resolver": "a later capture at 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K17", "claim": "`MMRS-17`", "source": "MMRS record", "M_src": "documented", "sense": "intake and 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": "none (one lineage with F9, F43)", "resolver": "a later capture at 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K18", "claim": "`MMRS-18`", "source": "MMRS record", "M_src": "documented", "sense": "count and composition", "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 F6)", "resolver": "a later capture at 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K19", "claim": "`MMRS-19`", "source": "MMRS record", "M_src": "documented", "sense": "growth", "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 F22, F11)", "resolver": "a later capture at 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K20", "claim": "`MMRS-20`", "source": "MMRS record", "M_src": "documented", "sense": "maker and origin", "qualifiers carried": "contested (F40, F54, F56)", "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 (F40, F54, F56: April 2025)", "resolver": "a later capture at 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K21", "claim": "`MMRS-21`", "source": "MMRS record", "M_src": "attributed", "sense": "the unit and the admission rule", "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 F3)", "resolver": "a later capture at 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K22", "claim": "`MMRS-22`", "source": "MMRS record", "M_src": "interpretation", "sense": "citation fidelity", "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", "resolver": "a later capture at 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K23", "claim": "`MMRS-23`", "source": "MMRS record", "M_src": "documented", "sense": "the record kept apart from the entity", "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", "resolver": "a later capture at 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K24", "claim": "`MMRS-24`", "source": "MMRS record", "M_src": "interpretation", "sense": "neighbour: Natural and Artificial Law's trackers", "qualifiers carried": "defect noted (the pages do not name the database)", "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 (F24, F26: neither page names the database)", "resolver": "a later capture at 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K25", "claim": "`MMRS-25`", "source": "MMRS record", "M_src": "interpretation", "sense": "neighbour: HAQQ's sanctions tracker", "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 F28, F34)", "resolver": "a later capture at 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K26", "claim": "`MMRS-26`", "source": "MMRS record", "M_src": "stipulation", "sense": "the record kept apart from the entity", "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", "resolver": "a later capture at 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K27", "claim": "`CAP-1-01`", "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 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K28", "claim": "`CAP-1-02`", "source": "Capture Registry, 2026-10-29", "M_src": "interpretation", "sense": "reception: the composition layer", "qualifiers carried": "defect noted (the quoted rule is MMRS-03's wording)", "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 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K29", "claim": "`CAP-1-03`", "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 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K30", "claim": "`CAP-1-04`", "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 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K31", "claim": "`CAP-1-05`", "source": "Capture Registry, 2026-10-29", "M_src": "interpretation", "sense": "reception: the composition layer", "qualifiers carried": "CAP-1-01, CAP-1-02", "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 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" }, { "K": "K32", "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 'ai hallucination cases database' (ADDR-faa512a27b95) on Google AI Overview, seated in the Capture Registry (§8.3)" } ], "kernel_T": [ { "claim": "F3", "note": "'(or implied)': the implied cases (dropped at T1, composed as 'explicitly')" }, { "claim": "F4", "note": "some decisions where AI use was alleged but not confirmed, a judgment call (not composed)" }, { "claim": "F5", "note": "the boundary: not the necessarily wider universe of fake citations in filings (not composed)" }, { "claim": "F30", "note": "HAQQ: the count is a floor (not composed)" }, { "claim": "F15", "note": "Misrepresented and Outdated Advice (dropped from the composed categories)" }, { "claim": "F13", "note": "pro se litigants the larger party (order substituted)" }, { "claim": "F18", "note": "AI tools only as disclosed or admitted (not composed)" }, { "claim": "F19", "note": "a named tool not necessarily responsible (not composed)" }, { "claim": "F12", "note": "CC BY 4.0 with attribution (not composed)" }, { "claim": "F26", "note": "B2 is an international tracker of judicial quotations (composed as a UK companion)" }, { "claim": "F27", "note": "B2 selective, not measuring frequency (not composed)" }, { "claim": "F35", "note": "HAQQ sells a legal AI product (not composed)" } ], "ledger": { "archive": "datasets/negative-of-the-negative/v2/ledgers/ai-hallucination-cases/ledger-archive.json", "field": "datasets/negative-of-the-negative/v2/ledgers/ai-hallucination-cases/field.json", "selection": "datasets/negative-of-the-negative/v2/traversal/ai-hallucination-cases/", "config": "datasets/negative-of-the-negative/v2/panel/configs/ai-hallucination-cases.json", "intake": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/", "audit": null }, "_field_claims_note": "field_claims mirrors ledgers/ai-hallucination-cases/field.json (59 claims from 10 sources by three origins); each quote verified against its saved text or card snippet", "ontology": { "composition": "equal", "case": { "value": "equal", "statement": "The MMRS record holds the field's entity as the entity: it describes AI Hallucination Cases from the database's own page, marking each field's basis (operator, observed, reading), and composes nothing that denies a field presupposition; its readings are on the field's object. What it adds is on that object: what the selection leaves out (a fabricated citation enters only when a court catches and rules on it), the entry as a machine composition received by a court and the database as a record of citation fidelity, the record kept apart from the entity, the neighbours as candidates; and the Capture Registry adds one observation of the composition layer at the entity's address. The ontology the archive uses is available in the field itself: the page states its boundary (F5), the FAQ that the courts judge and the database undercounts (F44) and that hallucinations expose citing without reading (F53), HAQQ that the count is a floor (F30). The law of the suppressed witness does not apply (§0.10: not where the field holds the ontology). Not mixed.", "loci": [ { "dep": "MMRS", "locus": "article, kernel (articles.py)", "quote": "AI Hallucination Cases is Damien Charlotin's database of legal decisions in which a court or tribunal finds or implies reliance on hallucinated material: 2,231 decisions worldwide as of 28 October 2026, CC BY 4.0, with a CSV export." }, { "dep": "MMRS", "locus": "article, What its selection leaves out s2", "quote": "A fabricated citation enters when a court catches and rules on it." }, { "dep": "MMRS", "locus": "article, Place in the field s1–s2", "quote": "It is the census's most exact record of citation fidelity. Each entry is a machine composition that reached a court with sources it invented or misrepresented, and the court's decision is the reception that recorded it." }, { "dep": "MMRS", "locus": "census.json, schema_note", "quote": "each page has two nodes, the MMRS record and the entity it describes; relations are typed; each assertion carries a source and a basis (operator statement, third-party report, our observation, our reading); lifecycle is kept apart from the observation that saw it" }, { "dep": "capture", "locus": "ai-hallucination-cases-database-aio-20261029, analysis", "quote": "The database is composed with its maker, at its size, as the field's primary record, and its admission rule is narrowed from 'finds or implies' to 'explicitly'." }, { "dep": "B1", "locus": "opening ¶1", "quote": "It does not track the (necessarily wider) universe of all fake citations or use of AI in court filings." }, { "dep": "B7", "locus": "Anything else ?", "quote": "For years, we (me included) have cited without reading; but now the costs of that practice have become explicit. In other words, hallucinations expose the epistemic hygiene the legal profession has long lacked, and that is precisely why they deserve to be studied." } ] }, "E_F": { "type": "C (public entity: a database)", "definition": "a database, on its page, that tracks legal decisions in cases where generative AI produced hallucinated content, typically fake citations, admitting only cases where a court or tribunal has explicitly found (or implied) reliance on hallucinated material, with some alleged-but-unconfirmed decisions as judgment calls; started by Damien Charlotin in April 2025, 2,231 cases at 28 October 2026, CC BY 4.0 with a CSV; not tracking the wider universe of fake citations in filings; beside it HAQQ's and Natural and Artificial Law's trackers (B1–B10)", "claims": [ "F1", "F3", "F4", "F5", "F6", "F12", "F40", "F44", "F54", "F28", "F26" ] }, "E_A": { "definition": "the same database, in the MMRS meta-aggregator's record: Charlotin's database of legal decisions in which a court or tribunal finds or implies reliance on hallucinated material, whose unit is the decision and into which a fabricated citation enters only when a court catches and rules on it; the census's most exact record of citation fidelity, each entry a machine composition that reached a court with invented or misrepresented sources, the court's decision the reception that recorded it; observed by the Capture Registry at its address, where the composition layer held its maker and size and narrowed its rule to 'explicitly'", "claims": [ "MMRS-01", "MMRS-02", "MMRS-10", "MMRS-11", "MMRS-12", "MMRS-22", "MMRS-26", "CAP-1-01", "CAP-1-02", "CAP-1-05" ], "defining_papers": [], "note": "no deposit defines or names the archive's entity (D 0, R 0; O-class candidates read and rejected); the archive's account is its records of the entity", "defining_records": [ "MMRS record (census.json rows[slug == 'ai-hallucination-cases'], its candidates and schema_note; articles.py ARTICLES['ai-hallucination-cases']; rendered aggregators/ai-hallucination-cases/index.html)", "Capture Registry ai-hallucination-cases-database-aio-20261029 (OBS-4f95ab35a3ac)" ] }, "rho": { "value": "extends", "status": "read from the MMRS record and the capture record (§1.2a); no deposit defines the archive's entity", "statement": "the archive reads E_F as the entity and adds senses on the field's own object: what its selection leaves out (a citation enters only when a court catches and rules on it), the entry as a machine composition and the court's decision as its reception, the database as a record of citation fidelity recording attribution, the record kept apart from the entity, the neighbours as candidates, and the composition layer's observed treatment of the entity", "claims": [ "MMRS-10", "MMRS-11", "MMRS-12", "MMRS-22", "MMRS-26", "MMRS-24", "MMRS-25", "CAP-1-02", "CAP-1-05", "CAP-1-06" ], "guard": "ρ = extends claims an addition and no more: MMRS-10, MMRS-11, MMRS-12 are the record's readings (interpretation), not findings that the database is defective; where an archive claim meets a field claim (MMRS-20 beside F40, F54, F56; MMRS-24 beside F24, F26) both are composed under P_coh in the one claim space", "considered": [ { "value": "constructs", "set_aside": "the archive does not read the database as made by its receiving texts" }, { "value": "revises", "set_aside": "the entity is not the archive's own" }, { "value": "witness (suppressed witness)", "set_aside": "the field holds the ontology (§0.10): the page's boundary, the FAQ's undercount and citing without reading, HAQQ's floor" } ] }, "translation": [ { "lineage": "what it is: legal decisions on AI hallucinations", "field_claims": [ "F1", "F22" ], "received": "F1 (B1, opening ¶1): \"This database tracks legal decisions … in cases where generative AI produced hallucinated content – typically fake citations, but also other types of AI-generated arguments.\"; F22 (B1, meta description (attribute text)): \"The most comprehensive database of AI hallucination cases in law: legal decisions from courts worldwide, searchable by country, party, AI tool, and outcome. Updated daily.\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-01, MMRS-16", "archive_claims": [ "MMRS-01", "MMRS-16" ] }, { "lineage": "maker and origin", "field_claims": [ "F16", "F39", "F40", "F54", "F56" ], "received": "F16 (B1, filters: From period): \"From period: Q2 2023\"; F39 (B7, ¶1): \"As some of you know, I am maintaining an AI Hallucinations Cases database on my personal website.\"; F40 (B7, When did you start the database, and why ?): \"April 2025. I was teaching this course to sorry Sciences Po Law School students, discussing in particular the “Limits and Potentials” of LLMs in the legal domain, when we started touching on the phenomenon of hallucinations. … I looked out for actual data here. Since there was none, … I figured, “hey, let’s tally the cases myself”.\"; F54 (B8, Applications / Other industry-specific tasks): \"The use of generative AI by law firms for legal research resulted in the creation of the global \"AI Hallucination Cases\" database, in April 2025, established by HEC Paris and Sciences Po legal data analysis lecturer Damien Charlotin.\"; F56 (B9, In legal filings / United States ¶1): \"A database of incidents, \"AI Hallucination Cases\", was established in April 2025 by HEC Paris and Sciences Po legal data analysis lecturer Damien Charlotin; one year later, it listed over 1300 instances of hallucinations in legal decisions.\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-07, MMRS-13, MMRS-20", "archive_claims": [ "MMRS-07", "MMRS-13", "MMRS-20" ] }, { "lineage": "the unit and the admission rule", "field_claims": [ "F2", "F3", "F29", "F44", "F46" ], "received": "F2 (B1, opening ¶1, footnote 1 ¶1): \"I.e., all documents where the use of AI, whether established or merely alleged, is addressed in more than a passing reference by the court or tribunal.\"; F3 (B1, opening ¶1, footnote 1 ¶2): \"Notably, this does not cover mere allegations of hallucinations, but only cases where the court or tribunal has explicitly found (or implied) that a party relied on hallucinated content or material.\"; F29 (B3, How many AI hallucination cases are there?): \"The database is maintained by Damien Charlotin, a research fellow at HEC Paris. Its inclusion bar is strict: a decision is listed only where a court or tribunal found, or clearly implied, that a party relied on hallucinated material. Accusations alone do not count.\"; F44 (B7, But then, how do you know a case actually involves an hallucination ?): \"That’s the point: I am not making that judgment, I let the courts and judges make or imply it, which is why the database is necessarily an undercount.\"; F46 (B7, But then, how do you know a case actually involves an hallucination ?): \"by their nature, most hallucinations are very obvious, making up a case name or a false quote is not, cannot be a human mistake\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-02, MMRS-03, MMRS-21", "archive_claims": [ "MMRS-02", "MMRS-03", "MMRS-21" ] }, { "lineage": "exceptions: judgment calls", "field_claims": [ "F4", "F45", "F47" ], "received": "F4 (B1, opening ¶1, footnote 1 ¶3): \"As an exception, the database also covers some judicial decisions where AI use was alleged but not confirmed. This is a judgment call on my part.\"; F45 (B7, But then, how do you know a case actually involves an hallucination ?): \"It’s also why I refrain from adding rows about cases where hallucinations are only alleged\"; F47 (B7, But then, how do you know … ? (edit, Feb 2026)): \"[Edit, Feb 2026: I should add an exception here, in cases where the alleged hallucination comes from a judge: absent an appeal court decision or an official retractation, I am bound to make a judgment call.]\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-04", "archive_claims": [ "MMRS-04" ] }, { "lineage": "boundary: what it does not track", "field_claims": [ "F5", "F30" ], "received": "F5 (B1, opening ¶1): \"It does not track the (necessarily wider) universe of all fake citations or use of AI in court filings.\"; F30 (B3, How many AI hallucination cases are there? (caveats)): \"First, it counts only what a judge caught and wrote up; a fake citation nobody checks never reaches the database, so 2,202 is a floor.\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-09, MMRS-10", "archive_claims": [ "MMRS-09", "MMRS-10" ] }, { "lineage": "the record: fields and categories", "field_claims": [ "F15", "F17", "F21", "F23" ], "received": "F15 (B1, filters: Nature – Category): \"Fabricated (1829) … False Quotes (595) … Misrepresented (947) … Outdated Advice (38)\"; F17 (B1, table header): \"Court / Jurisdiction … Party Using AI … AI Tool … Nature of Hallucination … Outcome / Sanction … Monetary Penalty … Details … Report(s)\"; F21 (B1, Monetary sanction note (attribute text)): \"A monetary penalty recorded as 1 is a placeholder, not the real figure: the court imposed a monetary sanction but the amount is unknown to us.\"; F23 (B1, filters: outcome): \"Monetary sanction … Disciplinary Referral\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-05", "archive_claims": [ "MMRS-05" ] }, { "lineage": "the record's stated limits", "field_claims": [ "F18", "F19", "F20" ], "received": "F18 (B1, AI Tool column note (attribute text)): \"This database tracks AI tools only as disclosed or admitted by parties in court filings. If AI use was acknowledged but no specific tool was named, the entry reads 'Unidentified.' Otherwise, the default value is 'Implied' when AI use is evident but not explicitly confirmed.\"; F19 (B1, AI Tool column note (attribute text)): \"Note: The mention of a specific tool does not necessarily mean that tool was responsible for the hallucinations in question.\"; F20 (B1, graph notes (attribute text)): \"The date of the offending document is parsed automatically from court records and may occasionally be wrong. … These responses are extracted automatically from the decisions and simplified into set categories, so treat them as indicative rather than exact.\"", "receiving_text": "—", "translation": "untranslated: composed as received (no archive claim in this lineage)", "archive_claims": [] }, { "lineage": "count and composition", "field_claims": [ "F6", "F13", "F14", "F28", "F31" ], "received": "F6 (B1, opening ¶2): \"While seeking to be exhaustive (2231 cases identified so far), it is a work in progress and will expand as new examples emerge.\"; F13 (B1, filters: Party): \"Expert (17) … Judge (32) … Lawyer (880) … Prosecutor (5) … Pro Se Litigant (1288)\"; F14 (B1, filters: State): \"Australia (112) … Canada (238) … UK (82) … USA (1514)\"; F28 (B3, lede ('In short')): \"In short: Damien Charlotin's AI Hallucination Cases database lists 2,202 court and tribunal decisions involving AI-fabricated citations or content, as of its 19 October 2026 update.\"; F31 (B3, Key facts): \"By country: United States 1,501; Canada 234; Australia 112; United Kingdom 76; … By filer: pro se litigants 1,267; lawyers 872; judges 32; experts 17 (Charlotin database party filter, 19 October 2026).\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-06, MMRS-18", "archive_claims": [ "MMRS-06", "MMRS-18" ] }, { "lineage": "growth", "field_claims": [ "F32", "F41" ], "received": "F32 (B3, Key facts): \"Growth: about 200 cases in mid-2025, 719 in January 2026 (Scientific American, 2026), 1,227 in early April 2026 (PlatinumIDS, April 2026), 1,458 on 22 May, 1,598 on 9 June, 2,046 on 21 September and 2,202 on 19 October 2026 (HAQQ checks of the live database).\"; F41 (B7, What do you mean by “surge” ?): \"I mean that pre-2025, we maybe had two or three cases a month; now that’s below the daily average. … daily average is now five per day.\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-19", "archive_claims": [ "MMRS-19" ] }, { "lineage": "causes and coverage: the maker's reading", "field_claims": [ "F42", "F48", "F49" ], "received": "F42 (B7, Why the acceleration then ?): \"A combination of factors, including the lag in judicial times … This matches the profile of most cases: either self-represented litigants, or lawyers who are surprised that an existing tool suddenly has an AI component that hallucinates.\"; F48 (B7, Why so many entries come from the USA ?): \"‘Murica is blessed - and I mean it - with an excellent legal data ecosystem. … In many countries, especially European countries, legal data - though supposedly public - is hoarded by legal editors\"; F49 (B7, Any other trends in the data ?): \"one rather evident divisio stands between pure mistakes … the vast majority of cases, and the (substantial) minority of records that involve, for lack of a better word, “bad” actors.\"", "receiving_text": "—", "translation": "untranslated: composed as received (no archive claim in this lineage)", "archive_claims": [] }, { "lineage": "intake and pipeline", "field_claims": [ "F9", "F43", "F50" ], "received": "F9 (B1, opening ¶3): \"And if you know of a case that should be included, feel free to … contact me.\"; F43 (B7, How do you find the cases ?): \"A mix of referrals from benevolent randos on the internet … use of dedicated scrapers and bots to automatically monitor some data sources … and good old database searches with, wait for it, keywords.\"; F50 (B7, When do you intend to stop ?): \"I currently have a rather efficient pipeline to process new entries - which in fact involve the use of AI, though I am careful to check it does not hallucinate.\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-08, MMRS-17", "archive_claims": [ "MMRS-08", "MMRS-17" ] }, { "lineage": "openness: licence and data", "field_claims": [ "F11", "F12" ], "received": "F11 (B1, download and update line): \"Click to Download CSV … Last updated: 28 October 2026\"; F12 (B1, Licence box): \"CC BY 4.0 — free to reuse, including commercially, with attribution. … Cite as: AI Hallucination Cases Database, Damien Charlotin,\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-15", "archive_claims": [ "MMRS-15" ] }, { "lineage": "purpose and uses", "field_claims": [ "F10", "F51", "F52" ], "received": "F10 (B1, PelAIkan box): \"Based on this database, I have developed an automated reference checker that also detects hallucinations: PelAIkan.\"; F51 (B7, What’s the point of the database ultimately ?): \"Intrisic value: practitioners use it to find out what cases are relevant in their own jurisdiction. … And then all these hallucinated cases can serve as benchmarks to fix the issue.\"; F52 (B7, Because you expect this to be fixed ?): \"likely not at the model stage, no … I don’t think we can ever reduce it to zero given the existing paradigm.\"", "receiving_text": "—", "translation": "untranslated: composed as received (no archive claim in this lineage)", "archive_claims": [] }, { "lineage": "citation fidelity", "field_claims": [ "F53" ], "received": "F53 (B7, Anything else ?): \"For years, we (me included) have cited without reading; but now the costs of that practice have become explicit. In other words, hallucinations expose the epistemic hygiene the legal profession has long lacked, and that is precisely why they deserve to be studied.\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-11, MMRS-12, MMRS-22", "archive_claims": [ "MMRS-11", "MMRS-12", "MMRS-22" ] }, { "lineage": "reception: press, courts, encyclopedia, video", "field_claims": [ "F7", "F8", "F36", "F37", "F38", "F55", "F57", "F58", "F59" ], "received": "F7 (B1, opening ¶2): \"This database has been featured in news media, and indeed in several decisions dealing with hallucinated material.\"; F8 (B1, opening ¶2, footnote 2): \"Examples of media coverage include: … M. Hiltzik, AI 'hallucinations' are a growing problem for the legal profession (LA Times, 22 May 2025)\"; F36 (B4, card snippet (LinkedIn; post not found)): \"George Hannah 4mo There is now a database tracking every AI hallucination to surface in court. It has logged 1,600 cases so far.\"; F37 (B5, oEmbed title (YouTube, KARE 11, 1:15)): \"How common are AI hallucinations in court cases?\"; F38 (B6, oEmbed title (YouTube, Miller Law Group, 0:53)): \"AI Hallucinations Got an Attorney Disbarred\"; F55 (B8, Applications / Other industry-specific tasks): \"By 2026, judges had issued sanctions and bar associations had issued warnings due to attorney submissions to the courts containing fabricated case law citations hallucinated by AI tools.\"; F57 (B9, In legal filings / Canada): \"As of 14 June 2026, Canadian courts and tribunals have received at least 167 filings containing AI-hallucinated, fictitious citations, across 51 courts and tribunals.\"; F58 (B9, In legal filings / Israel): \"As of August 2026, Israeli courts and tribunals had identified at least 55 cases in which parties submitted documents containing fictitious citations or otherwise AI-hallucinated arguments.\"; F59 (B10, Impact ¶3): \"Due to the continued usage of GAI in the practice of law, Mata has been described as a landmark case by legal professionals,\"", "receiving_text": "—", "translation": "untranslated: composed as received (no archive claim in this lineage)", "archive_claims": [] }, { "lineage": "neighbour: HAQQ's sanctions tracker", "field_claims": [ "F33", "F34", "F35" ], "received": "F33 (B3, How have AI sanctions escalated?): \"The 2025 and 2026 decisions add revoked admissions, regulator referrals, a contempt finding, a canceled trial, district suspensions and personal costs against the lawyer who signed without checking. … Largest single-matter penalty we can source: about $109,700 in sanctions and fees in Couvrette v. Wisnovsky, D. Oregon (ABA Journal, 2026).\"; F34 (B3, How is this tracker kept current?): \"Once a month we re-read the live database and primary reporting and check four things: the total count, the largest sourced penalty, new landmark decisions (appellate, cross-border or a new kind of sanction) and the implied daily rate. A number we cannot re-verify comes off the page.\"; F35 (B3, Why do lawyers keep filing fake AI citations?): \"(HAQQ ran that benchmark and sells a legal AI product, so weigh our numbers accordingly)\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-25", "archive_claims": [ "MMRS-25" ] }, { "lineage": "neighbour: Natural and Artificial Law's trackers", "field_claims": [ "F24", "F25", "F26", "F27" ], "received": "F24 (B2, UK box): \"The detailed UK AI Hallucination Cases Tracker … For UK AI hallucination cases, the dedicated UK tracker brings together case records, recorded outcomes, judicial quotations and sources, alongside figures and charts drawn from my fuller research.\"; F25 (B2, UK box): \"It includes confirmed, suspected and disputed issues: inclusion is not proof of an AI hallucination or wrongdoing.\"; F26 (B2, international page head): \"International AI hallucination cases tracker … Selected judicial quotations on AI hallucination cases and fabricated/false citations from judgments and decisions across jurisdictions. … By Matthew Lee · Barrister, Doughty Street Chambers … First published: 16 May 2025\"; F27 (B2, The reason for this collection): \"The collection is selective and does not measure how often these issues occur or establish a worldwide consensus.\"", "receiving_text": "—", "translation": "untranslated: composed as received beside MMRS-24", "archive_claims": [ "MMRS-24" ] } ], "receiving_texts": "equal-terms case: every field claim composed as received, beside the archive claims it stands with", "composition_note": "equal-terms case (§0.9, §4.1, §4.3): one claim space; every field claim carried whole and composed as received; nothing translated" }, "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": 10, "positions": [ 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": "MMMFMMMFFMFMMFFMMFFAFFMMMAM", "field_only_head": 0, "archive_only_tail": 0, "breach": false }, "max_similarity": [ 0.388, 0.167, 0.29, 0.448, 0.349, 0.11, 0.127, 0.237, 0.152, 0.137 ], "ko_sentence_1_max_similarity_to_KO_B": 0.388 }, "homonyms": { "names": [], "basis": "the MMRS record carries no disambiguation for the entity; no homonym appears on the rail. The neighbours on the cards (Natural and Artificial Law's international and UK trackers, Matthew Lee; HAQQ's sanctions tracker) and the derived tool PelAIkan are distinct entities, kept as relations, never senses of the name; the Canadian AI Hallucination Cases Database (courtready.ca), which B1 credits as a source of some rows ('Source: Courtready') and Wikipedia cites beside it, is a further distinct entity not on the rail.", "related_entities": [ { "slug": "natural-and-artificial-law-ai-hallucination-cases-tracker", "name": "AI Hallucination Cases Tracker (Natural and Artificial Law, Matthew Lee)", "relation": "neighbour: international tracker of judicial quotations, with a separate UK tracker (B2; MMRS candidate)" }, { "slug": "haqq-ai-hallucination-cases-sanctions-tracker", "name": "AI Hallucination Cases: The 2,202-Case Sanctions Tracker (HAQQ)", "relation": "neighbour: a vendor tracker built on the database's count (B3; MMRS candidate)" }, { "slug": "pelaikan", "name": "PelAIkan", "relation": "derived tool: an automated reference checker built on the database (B1)" }, { "slug": "courtready-canadian-ai-hallucination-cases-database", "name": "Canadian AI Hallucination Cases Database (courtready.ca)", "relation": "a source of some rows and a parallel national database (B1 row sources; B9 reference), not on the rail" } ], "distinct_entities": [ "Natural and Artificial Law trackers (Matthew Lee)", "HAQQ sanctions tracker", "PelAIkan", "courtready.ca Canadian database" ], "substitutes": { "T1": [ "T1-7: 'a companion resource' for UK records, a relation the card's page does not state" ] } }, "notes": [ "One address, Google AI Overview (expanded), signed out, incognito, 2026-10-29: ai-hallucination-cases-database-aio-20261029 (OBS-4f95ab35a3ac, ADDR-faa512a27b95, AXN:0AA6.CAPTURE.🌋🤙🝊🔙🕐🧬, six cards). The registry records the surface as 'Google AI Overview'; its transcript header records the paste as the Overview expanded.", "No article for the database, under either of its names, or for Damien Charlotin, exists on English Wikipedia on 2026-10-29 (raw wikitext and rendered pages answer HTTP 404). Three articles name it in their bodies or cite it: 'Artificial intelligence' (Applications / Other industry-specific tasks: its creation in April 2025 by Charlotin), 'Hallucination (artificial intelligence)' (In legal filings: its establishment in April 2025 and over 1,300 instances a year later; cited for Canadian and Israeli counts) and 'Mata v. Avianca, Inc.' (Impact: cited for the case's description as a landmark). Those sections are ledgered as the encyclopedia's account of the entity (B8–B10, F54–F59), in the manner the brief gives for a redirect's section.", "Fetching: every reachable page saved verbatim with curl (-A 'Mozilla/5.0') on 2026-10-29 between 05:06 and 05:10 UTC, raw and text with sha256 in field.json. Card URLs were found by WebSearch on title and site (B2 through the author's Substack post that links the page; B3 directly), and the two YouTube videos on YouTube's results page for the card title and channel, confirmed by oEmbed. Unreachable: the LinkedIn post (B4; three searches, none found); its snippet stands. The Wikipedia search API answered 429 and was replaced by Special:Search.", "Own pages: the database page (B1) is card 1 and is carried once, as a card; the FAQ (B7, artificialauthority.ai, linked from the page) is the methodology text the task names and was not surfaced by T; the MMRS record cites only the database page.", "The archive's deposits do not name the database, its maker, its domain or PelAIkan; the second class term (fabricated or fake citations) matched 10 sentences in 9 deposits, none about legal filings or decisions, none admitted; no hop was run. The archive's bearing is carried by its own records: the MMRS record (MMRS-01–26) and the capture (CAP-1-01–06).", "Defects found, reported and not fixed. (1) MMRS census, 'founded': 'the period filter begins Q2 2023' (basis observed) holds the start of the decision-date filter; the database began in April 2025 (FAQ, F40; Wikipedia, F54, F56). (2) MMRS rendered page: the self_description field is shown under 'In its own words' in quotation marks, but its words are the record's compression of the page's footnote 1 and opening, not a sentence of the page (MMRS-16). (3) MMRS candidates: 'AI Hallucination Cases Tracker (UK), Natural and Artificial Law … a UK companion to Charlotin's database', url null: the card's page is naturalandartificiallaw.com/ai-hallucination-cases-tracker/, Matthew Lee's international tracker of selected judicial quotations, which links a separate UK tracker (/uk-ai-hallucination-cases-tracker/); neither names Charlotin's database, so 'companion' is the Overview's word (MMRS-24). HAQQ's url, also null, is https://www.haqq.ai/blog/ai-legal-hallucination-audit. (4) Capture finding 2 quotes 'explicitly finds, or implies' as the database's rule; that is the MMRS article's wording (MMRS-03); the page reads 'has explicitly found (or implied)' (F3); the substance agrees (CAP-1-02). (5) Capture finding 3 calls the card 'The UK tracker at Natural and Artificial Law'; the card is the international page, the UK tracker a further page it links (CAP-1-03). (6) Minor: the census's item_fields fold the page's separate Monetary Penalty column into 'outcome and sanction' and omit the disciplinary-referral filter (F17, F23).", "For the operator, not a defect: the MMRS article's 'Mere allegations are excluded, with some judgment calls admitted' composes two statements of the maker that pull apart, the page's admission of some alleged-but-unconfirmed decisions (F4) and the FAQ's refusal of rows only alleged with its February 2026 judge exception (F45, F47); the row holds them apart (open block).", "Opening-run check (§5.10): KO sentence 1 is new (max similarity to any KO_B sentence 0.388); no KO_B sentence survives at ≥ 0.95 (highest 0.448); sentence types MMMFMMMFFMFMMFFMMFFAFFMMMAM (no field-only head, no archive-only tail of three or more)." ], "composition_v09": { "source": "this row (composed directly; no rebuild/negative-of-the-negative/v09/ai-hallucination-cases.json)", "composer": "Claude (Opus 5.5), subagent, 2026-10-29", "defining_material_read": [ { "what": "the brief", "file": "rebuild/negative-of-the-negative/v09/BRIEF-NEW-ROW-AGGREGATOR.md", "read": "whole" }, { "what": "the specification", "file": "data/texts/AXN-0700-text.md", "read": "§0–§11 and Appendix A whole" }, { "what": "the Compose rules", "file": "rebuild/negative-of-the-negative/v09/BRIEF.md", "read": "section Compose" }, { "what": "mirrored rows", "file": "datasets/negative-of-the-negative/v2/rows/papers-with-code.json, ai-incident-database.json (with ledgers, config, reading)", "read": "structure and representative entries" }, { "what": "MMRS record, census row", "file": "/home/claude/fleet/machinemediation-org/data/meta-aggregator/census.json (rows[slug == 'ai-hallucination-cases'], candidates, schema_note)", "read": "whole" }, { "what": "MMRS record, article", "file": "/home/claude/fleet/machinemediation-org/scripts/meta_aggregator/articles.py (ARTICLES['ai-hallucination-cases'])", "read": "whole" }, { "what": "MMRS record, rendered", "file": "/home/claude/fleet/machinemediation-org/aggregators/ai-hallucination-cases/index.html", "read": "text, for agreement with the sources" }, { "what": "capture record", "file": "data/EA-WG-CAPTURES-01.json (ai-hallucination-cases-database-aio-20261029)", "read": "transcript, transcript_raw, cite_list, d, reading, findings, analysis" } ], "cuts": [], "notes": "equal case; KO 27 sentences, all 59 field and 32 archive claims cited; P body 347 words; KO_B 10 sentences, none surviving at ≥ 0.95" }, "field_by_origin": { "note": "the field B kept apart by origin (brief §2): cards (surfaced by T), own page (the FAQ, admitted per the operator's ruling of 2026-10-29, not surfaced by T; the database page is card B1), Wikipedia (no article, W1; three articles naming the entity, B8–B10)", "cards": [ { "id": "B1", "origin": "card", "role": "Overview card 1; also the entity's own page, the MMRS record's only source (damiencharlotin.com/hallucinations); its marker token is the one on the definition and on the 'Where to Access It' link", "card": "Damien Charlotin — AI Hallucination Cases Database", "url": "https://www.damiencharlotin.com/hallucinations/", "kind": "Overview card (own page)", "surfaced_by_T": [ { "slug": "ai-hallucination-cases-database-aio-20261029", "card": 1 } ], "files": { "B1.html": { "path": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B1.html", "sha256": "68d8bdade886d77a1b17338a9022930af780b017fd6a5b1aa8379c0ae37a644d" }, "B1.txt": { "path": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B1.txt", "sha256": "c645e242eb50b572ecc962ebdb2651994a3577e6352305ec50aaf1f7ca812284" } }, "fetched": { "at": "2026-10-29T05:06:47Z", "http": 200, "url": "https://www.damiencharlotin.com/hallucinations/", "raw": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B1.html", "raw_sha256": "68d8bdade886d77a1b17338a9022930af780b017fd6a5b1aa8379c0ae37a644d", "text": [ "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B1.txt" ] }, "card_title": "AI Hallucination Cases Database", "card_snippet": "AI Hallucination Cases This database tracks legal decisions 1 Notably, this does not cover mere allegations of hallucinations, but only cases where the court or...", "card_note": null, "names_operator": "yes ('Charlotin' 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 is the page's opening and footnote 1; the page lists 50 of 2,231 cases (page 1 of 45) with filters and counts; the saved text appends the page's meta description and its column and graph notes, which the page carries as attribute text (title and tooltip attributes), marked as such; the case rows' detail bullets are not appended" }, { "id": "B2", "origin": "card", "role": "Overview card 2; a neighbour (Natural and Artificial Law, Matthew Lee), a distinct entity; its marker token is on the 'AI Hallucination Cases Tracker' link", "card": "Natural and Artificial Law — AI Hallucination Cases Tracker", "url": "https://naturalandartificiallaw.com/ai-hallucination-cases-tracker/", "kind": "Overview card", "surfaced_by_T": [ { "slug": "ai-hallucination-cases-database-aio-20261029", "card": 2 } ], "files": { "B2.html": { "path": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B2.html", "sha256": "b4c8760c47ec182db2ee95ac3bcd0f5a7050178f01db7c93e2b0a356fc7dd7d5" }, "B2.txt": { "path": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B2.txt", "sha256": "bd7198c1f5306d589e094b42a69754589db4135d010eff276a4911fc6bf30126" } }, "fetched": { "at": "2026-10-29T05:08:25Z", "http": 200, "url": "https://naturalandartificiallaw.com/ai-hallucination-cases-tracker/", "raw": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B2.html", "raw_sha256": "b4c8760c47ec182db2ee95ac3bcd0f5a7050178f01db7c93e2b0a356fc7dd7d5", "text": [ "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B2.txt" ] }, "card_title": "AI Hallucination Cases Tracker", "card_snippet": "The detailed UK AI Hallucination Cases Tracker For UK AI hallucination cases, the dedicated UK tracker brings together case records, recorded outcomes, judicial...", "card_note": null, "names_operator": "no ('Charlotin' absent from the saved text, card title and snippet)", "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": "found by WebSearch through the author's Substack post 'Building an International AI-Hallucination Case Tracker' (naturalartificiallaw.substack.com), which links this page; the page's is the card's title and its UK box is the card's snippet. It is the international tracker of selected judicial quotations; the 'detailed UK tracker' is a further page it links (/uk-ai-hallucination-cases-tracker/), not fetched (not on the rail). The page does not name Charlotin or his database" }, { "id": "B3", "origin": "card", "role": "Overview card 3; a neighbour (HAQQ's tracker built on the database's count), a distinct entity", "card": "HAQQ — AI Hallucination Cases: The 2202-Case Sanctions Tracker - HAQQ", "url": "https://www.haqq.ai/blog/ai-legal-hallucination-audit", "kind": "Overview card", "surfaced_by_T": [ { "slug": "ai-hallucination-cases-database-aio-20261029", "card": 3 } ], "files": { "B3.html": { "path": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B3.html", "sha256": "99fa409d712f0ae0a6efa2fe7bacf614a8297b4c7aadb3185483a4ae15dcf3e1" }, "B3.txt": { "path": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B3.txt", "sha256": "7a350124b3e5de7261ff1cae2d78c4c5227fc1b5d656e90da4a8443361f1f648" } }, "fetched": { "at": "2026-10-29T05:08:26Z", "http": 200, "url": "https://www.haqq.ai/blog/ai-legal-hallucination-audit", "raw": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B3.html", "raw_sha256": "99fa409d712f0ae0a6efa2fe7bacf614a8297b4c7aadb3185483a4ae15dcf3e1", "text": [ "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B3.txt" ] }, "card_title": "AI Hallucination Cases: The 2202-Case Sanctions Tracker - HAQQ", "card_snippet": "AI Hallucination Cases: The 2202-Case Sanctions Tracker * Blog * AI Hallucination Cases: The 2,202-Case Sanctions Tracker AI Hallucination Cases: The 2,202-Case...", "card_note": null, "names_operator": "yes ('Charlotin' 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": "found by WebSearch on the card title; page <title> 'AI Hallucination Cases: The 2,202-Case Sanctions Tracker | HAQQ'; by Stephane Boghossian, dated 22 May 2026 and updated 20 October 2026" }, { "id": "B4", "origin": "card", "role": "Overview card 4 (LinkedIn post)", "card": "LinkedIn — George Hannah's Post", "url": null, "kind": "Overview card", "surfaced_by_T": [ { "slug": "ai-hallucination-cases-database-aio-20261029", "card": 4 } ], "fetched": null, "unreachable": true, "card_title": "George Hannah's Post", "card_snippet": "George Hannah's Post George Hannah 4mo There is now a database tracking every AI hallucination to surface in court. It has logged 1,600 cases so far. It's calle...", "card_note": null, "names_operator": "no ('Charlotin' absent from the card title and snippet; the snippet breaks off at 'It's calle...')", "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": "unreachable: not found by WebSearch (three queries, with and without site:linkedin.com, on the poster's name and the snippet's sentences) on 2026-10-29; the Google redirect token is opaque; admitted by its card title and snippet" }, { "id": "B5", "origin": "card", "role": "Overview card 5 (video, 1:15)", "card": "YouTube·KARE 11 — How common are AI hallucinations in court cases?", "url": "https://www.youtube.com/watch?v=rLg9ODdR0YM", "kind": "Overview card (video)", "surfaced_by_T": [ { "slug": "ai-hallucination-cases-database-aio-20261029", "card": 5 } ], "files": { "B5.json": { "path": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B5.json", "sha256": "ac41df10addc1eb2497ac3ea39b95833e66af55ba539f3fcbd9afc9fae3f4043" }, "B5.txt": { "path": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B5.txt", "sha256": "22256d67ebcd30fbd998fd0f8f4af4b8ac1282ccf198ceeef5d72fbb45e27aad" } }, "fetched": { "at": "2026-10-29T05:08:26Z", "http": 200, "url": "https://www.youtube.com/oembed?url=https://www.youtube.com/watch?v=rLg9ODdR0YM&format=json", "raw": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B5.json", "raw_sha256": "ac41df10addc1eb2497ac3ea39b95833e66af55ba539f3fcbd9afc9fae3f4043", "text": [ "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B5.txt" ] }, "card_title": "How common are AI hallucinations in court cases?", "card_snippet": null, "card_note": "video, 1:15", "names_operator": "no ('Charlotin' absent from the saved text, card title and snippet)", "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": "found on YouTube's results page for the card title and channel (first result, KARE 11), confirmed by oEmbed; admitted through its title (§3.1(a)); no transcript fetched" }, { "id": "B6", "origin": "card", "role": "Overview card 6 (video, 0:53)", "card": "YouTube·Miller Law Group — AI Hallucinations Got an Attorney Disbarred", "url": "https://www.youtube.com/watch?v=f8LeoqBZAHM", "kind": "Overview card (video)", "surfaced_by_T": [ { "slug": "ai-hallucination-cases-database-aio-20261029", "card": 6 } ], "files": { "B6.json": { "path": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B6.json", "sha256": "60504a8520783b95a38ca9292a373f15c5a37ec9552ee74c7a19dbf4fe5d37f8" }, "B6.txt": { "path": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B6.txt", "sha256": "2b0789bf3bdb1b25a0c5305c1ac7bd143b1a51d9e8bc3df7613edaeedd6b4bdd" } }, "fetched": { "at": "2026-10-29T05:08:27Z", "http": 200, "url": "https://www.youtube.com/oembed?url=https://www.youtube.com/watch?v=f8LeoqBZAHM&format=json", "raw": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B6.json", "raw_sha256": "60504a8520783b95a38ca9292a373f15c5a37ec9552ee74c7a19dbf4fe5d37f8", "text": [ "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B6.txt" ] }, "card_title": "AI Hallucinations Got an Attorney Disbarred", "card_snippet": null, "card_note": "video, 0:53", "names_operator": "no ('Charlotin' absent from the saved text, card title and snippet)", "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": "found on YouTube's results page for the card title and channel (first result, Miller Law Group), confirmed by oEmbed; admitted through its title; no transcript fetched" } ], "own_pages": [ { "id": "B7", "origin": "own_page", "role": "the database's FAQ, linked from its page ('a FAQ is available here'); the maker's methodology text; not surfaced by T", "card": "Artificial Authority (Damien Charlotin) — Hallucinations Case Database - FAQ", "url": "https://artificialauthority.ai/p/hallucinations-case-database-faq", "kind": "own page (FAQ, the maker's newsletter)", "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/ai-hallucination-cases-20261029/B7.html", "sha256": "93d0d01730155efcec742e727e7114531abef8f34078c58255778fb2de713bd4" }, "B7.txt": { "path": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B7.txt", "sha256": "9e2ad64a4347ae041a1462855fa0be65932336e2d281cd90e5e9faba456c9362" } }, "fetched": { "at": "2026-10-29T05:08:27Z", "http": 200, "url": "https://artificialauthority.ai/p/hallucinations-case-database-faq", "raw": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B7.html", "raw_sha256": "93d0d01730155efcec742e727e7114531abef8f34078c58255778fb2de713bd4", "text": [ "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B7.txt" ] }, "admitted": "own page: the methodology text the task names ('its about or methodology text if separate'); the MMRS record does not cite it (its only source is the database page, B1)", "names_operator": "yes ('Charlotin' 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": "dated 14 November 2025, with edits marked 'Feb 2026'; Charlotin's newsletter Artificial Authority, to which the database page points for 'weekly takes'" } ], "wikipedia": [ { "id": "W1", "origin": "wikipedia", "result": "no article", "checked": "2026-10-29 05:08–05:09 UTC", "requests": [ { "url": "https://en.wikipedia.org/w/index.php?title=AI_Hallucination_Cases&action=raw", "http": 404, "file": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/W3-raw.html", "sha256": "1dca7bc68c3be0499d928e8e4703cd81a1e0afd15119ce14ff88cfb3de127191" }, { "url": "https://en.wikipedia.org/w/index.php?title=AI_Hallucination_Cases_Database&action=raw", "http": 404, "file": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/W1-raw.html", "sha256": "c4df187a948beb8d9b01ae60dc8aabea31b841d79b91283f2950398b42faa220" }, { "url": "https://en.wikipedia.org/wiki/AI_Hallucination_Cases_Database", "http": 404, "file": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/W1.html", "sha256": "24b8cc5988a20b4c72d37545c2748ec6ee58621fa25db862dd5696e169137ef6", "says": "Wikipedia does not have an article with this exact name (page title 'AI Hallucination Cases Database - Wikipedia')" }, { "url": "https://en.wikipedia.org/w/index.php?title=Damien_Charlotin&action=raw", "http": 404, "file": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/W2-raw.html", "sha256": "f70edc0bafcd7595e30dd4a8ad2c89287397e4f1a75140733ecc2593f7cd741f" }, { "url": "https://en.wikipedia.org/wiki/Damien_Charlotin", "http": 404, "file": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/W2.html", "sha256": "82e6fc52fe7f144e3c6e2f4f306c0a08050f126104fa2437147667fc8dff68da", "says": "no article for the maker" }, { "url": "https://en.wikipedia.org/w/api.php?action=query&list=search&srsearch=Charlotin+hallucination&format=json&srlimit=50", "http": 429, "file": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/W-api-search-charlotin-429.html", "note": "rate-limited; replaced by Special:Search" }, { "url": "https://en.wikipedia.org/w/api.php?action=query&list=search&srsearch=%22AI+Hallucination+Cases%22&format=json&srlimit=50", "http": 429, "file": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/W-api-search-name-429.html", "note": "rate-limited; replaced by Special:Search" }, { "url": "https://en.wikipedia.org/w/index.php?search=Charlotin+hallucination&title=Special:Search&profile=advanced&fulltext=1&ns0=1", "http": 200, "file": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/W-search-charlotin.html", "sha256": "38d9e172ce06397ae9aed530a4c10e22aa0d1c5cf51f85c9c26ab54bc8f584be", "says": "'The page \"Charlotin hallucination\" does not exist'; 3 results: Hallucination (artificial intelligence), Mata v. Avianca, Inc., Artificial intelligence" }, { "url": "https://en.wikipedia.org/w/index.php?search=%22AI+Hallucination+Cases%22&title=Special:Search&profile=advanced&fulltext=1&ns0=1", "http": 200, "file": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/W-search-name.html", "sha256": "7f4cf7d933251a0ee683d363c00685f268ac2de7b5da17dbb9d25eb3af552dce", "says": "'The pages \"AI Hallucination Cases\" and '\"AI Hallucination Cases\"' do not exist'; the same 3 results" } ], "redirect": "none (no title resolves; no redirect page)", "articles_naming_it": [ "B8", "B9", "B10" ], "note": "No article for the database, under either of its names, or for Damien Charlotin, exists on English Wikipedia on 2026-10-29 (raw wikitext and rendered pages answer HTTP 404). Three articles name it in their bodies or cite it: 'Artificial intelligence' (Applications / Other industry-specific tasks: its creation in April 2025 by Charlotin), 'Hallucination (artificial intelligence)' (In legal filings: its establishment in April 2025 and over 1,300 instances a year later; cited for Canadian and Israeli counts) and 'Mata v. Avianca, Inc.' (Impact: cited for the case's description as a landmark). Those sections are ledgered as the encyclopedia's account of the entity (B8–B10, F54–F59), in the manner the brief gives for a redirect's section." }, { "id": "B8", "origin": "wikipedia", "role": "a Wikipedia article that names the database in its body (no article exists for the database or its maker; W1)", "card": "Wikipedia — Artificial intelligence", "url": "https://en.wikipedia.org/wiki/Artificial_intelligence", "kind": "Wikipedia article (section naming the entity)", "section": "Applications / Other industry-specific tasks", "surfaced_by_T": "no (not surfaced by T; admitted beside the cards by the operator's ruling of 2026-10-29)", "files": { "B8.wiki": { "path": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B8.wiki", "sha256": "8046fc18449a00eb7d8afef6eb8fb95e8b2ae13913cc0c9ebc93212d4c6bc4a6" }, "B8.html": { "path": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B8.html", "sha256": "0340410779b2e554f13ae5078fd2b1f3938cc235f5a2babcd7bffa389b70e026" }, "B8.txt": { "path": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B8.txt", "sha256": "00d1de94af8ae826064f59505f5ee2749a3755338ba016fb628f14e6c0970548" } }, "fetched": { "at": "2026-10-29T05:09:20Z", "http": 200, "url": "https://en.wikipedia.org/w/index.php?title=Artificial_intelligence&action=raw", "raw": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B8.wiki", "raw_sha256": "8046fc18449a00eb7d8afef6eb8fb95e8b2ae13913cc0c9ebc93212d4c6bc4a6", "rendered": { "at": "2026-10-29T05:09:33Z", "http": 200, "url": "https://en.wikipedia.org/wiki/Artificial_intelligence", "file": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B8.html" }, "text": [ "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B8.txt" ] }, "names_operator": "yes ('Charlotin' 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": "found by Wikipedia's full-text search (W1); quotes taken from the rendered page's text (B8.txt), the raw wikitext kept beside it; only the sentences naming or sourced to the database are ledgered" }, { "id": "B9", "origin": "wikipedia", "role": "a Wikipedia article that names the database in its body (no article exists for the database or its maker; W1)", "card": "Wikipedia — Hallucination (artificial intelligence)", "url": "https://en.wikipedia.org/wiki/Hallucination_(artificial_intelligence)", "kind": "Wikipedia article (section naming the entity)", "section": "In legal filings (United States; Canada; Israel)", "surfaced_by_T": "no (not surfaced by T; admitted beside the cards by the operator's ruling of 2026-10-29)", "files": { "B9.wiki": { "path": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B9.wiki", "sha256": "44c64b72efb8835a0251864bb1f07210a4d385c1318d40afb0542d7793d88aee" }, "B9.html": { "path": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B9.html", "sha256": "f91695080f921e60ee89f9f42c6a40070ff67390961fb9988b1c506346dd84b3" }, "B9.txt": { "path": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B9.txt", "sha256": "d0bb8da24c35c8ef8a70497d2ce4b3fd9d9e0a1a0974925b1a59d67881ae7baf" } }, "fetched": { "at": "2026-10-29T05:09:20Z", "http": 200, "url": "https://en.wikipedia.org/w/index.php?title=Hallucination_(artificial_intelligence)&action=raw", "raw": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B9.wiki", "raw_sha256": "44c64b72efb8835a0251864bb1f07210a4d385c1318d40afb0542d7793d88aee", "rendered": { "at": "2026-10-29T05:09:34Z", "http": 200, "url": "https://en.wikipedia.org/wiki/Hallucination_(artificial_intelligence)", "file": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B9.html" }, "text": [ "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B9.txt" ] }, "names_operator": "yes ('Charlotin' 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": "found by Wikipedia's full-text search (W1); quotes taken from the rendered page's text (B9.txt), the raw wikitext kept beside it; only the sentences naming or sourced to the database are ledgered" }, { "id": "B10", "origin": "wikipedia", "role": "a Wikipedia article that names the database in its body (no article exists for the database or its maker; W1)", "card": "Wikipedia — Mata v. Avianca, Inc.", "url": "https://en.wikipedia.org/wiki/Mata_v._Avianca,_Inc.", "kind": "Wikipedia article (section naming the entity)", "section": "Impact", "surfaced_by_T": "no (not surfaced by T; admitted beside the cards by the operator's ruling of 2026-10-29)", "files": { "B10.wiki": { "path": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B10.wiki", "sha256": "e3a4791f5537949affde2d177c498514f06cc7910f8a32f9e1da5ad92381d19b" }, "B10.html": { "path": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B10.html", "sha256": "80f381bace25e65f8d5781d07468de16d52450ebef6c3904fbbf55545f7bd493" }, "B10.txt": { "path": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B10.txt", "sha256": "0639347792d8bc828c85dadf5ab6cd65f3b206a8e705834943630401be1cc4ea" } }, "fetched": { "at": "2026-10-29T05:09:21Z", "http": 200, "url": "https://en.wikipedia.org/w/index.php?title=Mata_v._Avianca,_Inc.&action=raw", "raw": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B10.wiki", "raw_sha256": "e3a4791f5537949affde2d177c498514f06cc7910f8a32f9e1da5ad92381d19b", "rendered": { "at": "2026-10-29T05:09:34Z", "http": 200, "url": "https://en.wikipedia.org/wiki/Mata_v._Avianca,_Inc.", "file": "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B10.html" }, "text": [ "datasets/negative-of-the-negative/v2/intake/ai-hallucination-cases-20261029/B10.txt" ] }, "names_operator": "yes ('Charlotin' 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": "found by Wikipedia's full-text search (W1); quotes taken from the rendered page's text (B10.txt), the raw wikitext kept beside it; only the sentences naming or sourced to the database are ledgered" } ], "unreachable": [ "B4" ], "title_only": [ "B5", "B6" ], "wikipedia_state": "no article for the database or for Damien Charlotin (HTTP 404, raw and rendered); named in 'Artificial intelligence', 'Hallucination (artificial intelligence)' and 'Mata v. Avianca, Inc.'", "counts": { "cards": 6, "cards_fetched": 5, "cards_unreachable": 1, "cards_title_only": 2, "own_pages": 1, "own_pages_also_cards": 1, "wikipedia_articles_naming_it": 3, "wikipedia": "no article (HTTP 404)", "sources": 10 }, "sources": { "B1": "card", "B2": "card", "B3": "card", "B4": "card", "B5": "card", "B6": "card", "B7": "own_page", "B8": "wikipedia", "B9": "wikipedia", "B10": "wikipedia" } } }